INDEX

Explanations

sentences where someone is being asked questions

oai_token-act-pair · gpt-3.5-turbo

instances of the word "asked."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

herence

-0.66

Est

-0.64

Fit

-0.64

Ĥ¬

-0.64

ources

-0.62

Inc

-0.60

Liber

-0.60

Zone

-0.60

 mars

-0.60

 Confeder

-0.59

POSITIVE LOGITS

 questions

1.06

 repeatedly

0.88

 question

0.85

naires

0.85

 asked

0.82

naire

0.82

 rhet

0.81

 forgiveness

0.78

 permission

0.76

 asks

0.76

Activations Density 0.036%

sentences where someone is being asked questions

instances of the word "asked."

No Comments

No Known Activations