INDEX

Explanations

instances where a specific task or action is suggested as being appropriate or beneficial

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate specific conditions or situations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

gem

-0.72

agin

-0.72

edly

-0.67

harm

-0.65

ãĤ«

-0.64

omers

-0.64

Bas

-0.63

rolet

-0.63

athom

-0.61

hash

-0.60

POSITIVE LOGITS

soever

1.36

irlf

0.87

 confronted

0.80

*/(

0.78

 pressed

0.75

 faced

0.73

 asked

0.73

 comparing

0.70

 evaluating

0.70

 subjected

0.69

Activations Density 0.134%

instances where a specific task or action is suggested as being appropriate or beneficial

phrases that indicate specific conditions or situations

No Comments

No Known Activations