INDEX

Explanations

locations or areas

oai_token-act-pair · gpt-3.5-turbo

references to various types of spaces, particularly those related to public and community environments

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

 ransom

-0.69

 blackmail

-0.68

Uz

-0.63

ITH

-0.62

 neurot

-0.62

 vengeance

-0.62

sonian

-0.60

 Edmund

-0.59

ocese

-0.58

 likeness

-0.58

POSITIVE LOGITS

 spaces

1.09

hips

1.00

 Spaces

0.96

hops

0.89

uits

0.89

pace

0.88

paces

0.87

 stalls

0.83

adia

0.83

ility

0.80

Activations Density 0.038%

locations or areas

references to various types of spaces, particularly those related to public and community environments

No Comments

No Known Activations