INDEX

Explanations

phrases indicating a suggested action

oai_token-act-pair · gpt-3.5-turbo

phrases or clauses expressing purpose or intention, typically indicated by the phrase "so that."

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

ARK

-0.68

ãĤ´ãĥ³

-0.62

 Exodus

-0.61

PRESS

-0.60

ART

-0.60

DAQ

-0.59

Plus

-0.59

AND

-0.58

arag

-0.58

ãĥ¡

-0.57

POSITIVE LOGITS

uces

0.75

 they

0.74

 whoever

0.69

 nobody

0.68

 maxim

0.68

ndra

0.67

 persists

0.66

ancial

0.64

we

0.62

pload

0.62

Activations Density 0.052%

phrases indicating a suggested action

phrases or clauses expressing purpose or intention, typically indicated by the phrase "so that."

No Comments

No Known Activations