INDEX

Explanations

sets of instructions with a specific action followed by a reason or consequence

oai_token-act-pair · gpt-3.5-turbo

conditional phrases that imply consequences or results

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

 Mens

-0.69

inch

-0.69

 understatement

-0.69

 glances

-0.66

 gallery

-0.64

 praise

-0.62

question

-0.60

Nay

-0.59

ropolitan

-0.58

 disbelief

-0.58

POSITIVE LOGITS

oner

0.98

bered

0.98

arer

0.95

apy

0.90

letes

0.90

othe

0.86

oths

0.84

aps

0.82

ooo

0.82

fter

0.81

Activations Density 0.075%

sets of instructions with a specific action followed by a reason or consequence

conditional phrases that imply consequences or results

No Comments

No Known Activations