INDEX

Explanations

verbs that denote some form of action or activity, often related to sharing, suggesting, discussing, testing, asking, showing, providing, editing, filming, viewing, recommending, uploading, warning, making, and revealing

oai_token-act-pair · gpt-3.5-turbo

verbs related to actions that have been performed or completed

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.6.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

rouse

-0.66

rients

-0.61

chal

-0.60

ears

-0.57

 offending

-0.55

arist

-0.55

 Latter

-0.54

 pops

-0.53

 Origin

-0.52

rit

-0.52

POSITIVE LOGITS

 lately

0.82

 since

0.80

by

0.76

 psychiat

0.75

 elsewhere

0.74

 unanimously

0.70

otyp

0.69

Ļ½

0.69

iasco

0.69

IFIED

0.69

Activations Density 0.153%

verbs that denote some form of action or activity, often related to sharing, suggesting, discussing, testing, asking, showing, providing, editing, filming, viewing, recommending, uploading, warning, making, and revealing

verbs related to actions that have been performed or completed

No Comments

No Known Activations