INDEX

Explanations

instances of the word "interview"

oai_token-act-pair · gpt-3.5-turbo

references to interviews

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

 Goods

-0.78

ple

-0.76

ciplinary

-0.69

wagen

-0.66

benefit

-0.65

profits

-0.63

hover

-0.63

 crest

-0.63

ignt

-0.62

warm

-0.62

POSITIVE LOGITS

 interview

0.94

 interviews

0.85

ees

0.78

 aired

0.77

athon

0.73

 conducted

0.72

reon

0.72

VO

0.72

 newsp

0.71

Interview

0.70

Activations Density 0.014%

instances of the word "interview"

references to interviews

No Comments

No Known Activations