INDEX

Explanations

legal and judicial terms

oai_token-act-pair · gpt-3.5-turbo

terms related to legal actions and proceedings

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

ceive

-0.66

thood

-0.59

DonaldTrump

-0.58

arate

-0.57

avement

-0.57

tnc

-0.57

ushima

-0.55

Semitism

-0.55

cale

-0.54

NetMessage

-0.54

POSITIVE LOGITS

 consisted

0.66

iest

0.65

 Ambro

0.62

 consists

0.61

 portion

0.60

 however

0.60

 factor

0.59

ultimate

0.59

 succeeded

0.59

 aspect

0.58

Activations Density 1.157%

legal and judicial terms

terms related to legal actions and proceedings

No Comments

No Known Activations