INDEX

Explanations

responses to various stimuli or situations

oai_token-act-pair · gpt-3.5-turbo

mentions of various types of responses

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Zot

-0.73

cutting

-0.71

dar

-0.65

ramer

-0.64

rome

-0.64

fall

-0.63

cin

-0.63

akia

-0.62

BALL

-0.60

ondo

-0.59

POSITIVE LOGITS

 thereto

0.98

aries

0.88

 response

0.87

ivated

0.87

iveness

0.85

naires

0.83

 elic

0.80

 Response

0.77

to

0.76

 responses

0.71

Activations Density 0.039%

responses to various stimuli or situations

mentions of various types of responses

No Comments

No Known Activations