INDEX

Explanations

words related to legal matters or formal documents

oai_token-act-pair · gpt-3.5-turbo

linguistic structures related to the suffixes or endings of words

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

 Rabbit

-0.72

 Cheong

-0.69

 Lines

-0.66

 Werewolf

-0.66

Hy

-0.65

 Scorp

-0.64

 bitterly

-0.63

 Isles

-0.62

 Pavilion

-0.62

 cracked

-0.62

POSITIVE LOGITS

utive

0.92

cially

0.91

iously

0.90

utor

0.87

andise

0.85

hement

0.85

ote

0.83

inarily

0.82

ciating

0.81

icit

0.81

Activations Density 0.170%

words related to legal matters or formal documents

linguistic structures related to the suffixes or endings of words

No Comments

No Known Activations