INDEX

Explanations

organizations or entities with the term "Association" in their name

oai_token-act-pair · gpt-3.5-turbo

references to various associations

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

lasses

-0.85

gone

-0.74

tro

-0.73

lio

-0.71

stocks

-0.70

lass

-0.69

mr

-0.66

=-=-=-=-=-=-=-=-

-0.66

trap

-0.64

nir

-0.64

POSITIVE LOGITS

UTH

0.88

uthor

0.85

 Association

0.84

SPA

0.76

 Agreement

0.75

auga

0.75

 membership

0.74

 Network

0.74

 Board

0.74

 Federation

0.74

Activations Density 0.019%

organizations or entities with the term "Association" in their name

references to various associations

No Comments

No Known Activations