INDEX

Explanations

specific terms or provisions mentioned in legal or legislative contexts

oai_token-act-pair · gpt-3.5-turbo

specific terminologies related to legal, procedural, or regulatory contexts

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

bats

-0.79

idents

-0.76

ateurs

-0.72

ositories

-0.70

amaz

-0.69

visors

-0.69

osponsors

-0.69

endars

-0.68

llah

-0.67

 finances

-0.67

POSITIVE LOGITS

 called

0.96

 whereby

0.82

 Called

0.80

 named

0.79

 nicknamed

0.75

nown

0.75

 dubbed

0.70

 titled

0.70

 pecul

0.70

 whose

0.69

Activations Density 0.426%

specific terms or provisions mentioned in legal or legislative contexts

specific terminologies related to legal, procedural, or regulatory contexts

No Comments

No Known Activations