INDEX

Explanations

disclosures or statements made by individuals

oai_token-act-pair · gpt-3.5-turbo

instances of the word "said" and phrases introducing reported speech or statements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

JV

-0.75

Ranked

-0.67

 Pont

-0.66

-0.65

Cyp

-0.64

ctors

-0.64

-0.63

Commerce

-0.62

MJ

-0.61

ãĢ

-0.61

POSITIVE LOGITS

 repeatedly

1.00

avering

0.95

 since

0.90

 unequivocally

0.89

querque

0.81

 stren

0.75

 numerous

0.73

 extensively

0.72

ittal

0.71

ffield

0.70

Activations Density 0.140%

disclosures or statements made by individuals

instances of the word "said" and phrases introducing reported speech or statements

No Comments

No Known Activations