INDEX

Explanations

the word "Coll" with varying levels of activation

oai_token-act-pair · gpt-3.5-turbo

references to "Colliers" or other related terms associated with that name

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

¥µ

-0.81

oshenko

-0.78

visor

-0.76

thing

-0.74

PDATE

-0.73

isSpecialOrderable

-0.71

FORE

-0.71

ItemImage

-0.70

WARE

-0.69

fulness

-0.68

POSITIVE LOGITS

ateral

1.12

apsed

1.02

apse

0.99

ision

0.97

isions

0.96

ossus

0.92

oqu

0.90

iery

0.89

onel

0.88

ouri

0.86

Activations Density 0.009%

the word "Coll" with varying levels of activation

references to "Colliers" or other related terms associated with that name

No Comments

No Known Activations