INDEX

Explanations

words related to legal actions or decisions

oai_token-act-pair · gpt-3.5-turbo

references to legal orders and directives

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

ãĤ©

-0.86

ãĤ§

-0.78

Unity

-0.76

ãĤ¤ãĥĪ

-0.74

 Volks

-0.73

dden

-0.73

 Sporting

-0.70

rum

-0.70

 Munich

-0.69

 Miliband

-0.69

POSITIVE LOGITS

lies

1.04

 warrants

0.91

 orders

0.86

 decree

0.86

 authorizing

0.83

 warrant

0.83

etary

0.83

 forb

0.82

 prohibiting

0.82

 issued

0.81

Activations Density 0.024%

words related to legal actions or decisions

references to legal orders and directives

No Comments

No Known Activations