INDEX

Explanations

mentions of disapproval or criticism

oai_token-act-pair · gpt-3.5-turbo

a specific character or symbol in the text

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

 deed

-0.69

 tabl

-0.69

 Norn

-0.68

 apes

-0.68

 Slug

-0.67

 prefrontal

-0.65

 telesc

-0.64

 pleasures

-0.64

 condem

-0.63

 Directorate

-0.62

POSITIVE LOGITS

ï¸ı

1.09

âĶĢâĶĢ

0.95

ternity

0.92

lean

0.92

ever

0.91

âĸł

0.87

\-

0.85

conom

0.84

Â·Â·

0.83

very

0.83

Activations Density 0.200%

mentions of disapproval or criticism

a specific character or symbol in the text

No Comments

No Known Activations