INDEX

Explanations

references to specific names and terms related to legal contexts or notable individuals

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

references to geographical locations

oai_token-act-pair · gpt-3.5-turbo

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

selves

-0.85

rift

-0.83

ithing

-0.78

quickShipAvailable

-0.77

owered

-0.75

erous

-0.74

ebted

-0.73

roud

-0.73

renches

-0.72

istan

-0.72

POSITIVE LOGITS

fruit

0.94

vine

0.92

burg

0.80

PER

0.75

hene

0.71

Wax

0.70

 juices

0.69

 Corpus

0.68

CLE

0.66

FTA

0.66

Activations Density 0.018%

references to specific names and terms related to legal contexts or notable individuals

references to geographical locations

No Comments

No Known Activations