INDEX

Explanations

phrases related to legal cases or situations

oai_token-act-pair · gpt-3.5-turbo

references to "case," particularly in legal or specific 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

 mathemat

-0.79

 destro

-0.79

etsk

-0.78

pez

-0.74

 livest

-0.73

¥ŀ

-0.73

emouth

-0.72

autical

-0.72

ower

-0.71

nesday

-0.71

POSITIVE LOGITS

cases

0.93

backs

0.80

case

0.78

cmp

0.76

back

0.70

Kas

0.70

suit

0.68

++++++++++++++++

0.65

xia

0.65

esan

0.65

Activations Density 0.036%

phrases related to legal cases or situations

references to "case," particularly in legal or specific contexts

No Comments

No Known Activations