INDEX

Explanations

terms related to authorization and organization activities

oai_token-act-pair · gpt-3.5-turbo

abbreviations or acronyms typically associated with organizations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

prise

-0.95

*/(

-0.74

 stump

-0.73

lopp

-0.73

icide

-0.70

wash

-0.70

parency

-0.69

verting

-0.67

prises

-0.67

itals

-0.67

POSITIVE LOGITS

BuyableInstoreAndOnline

0.86

ansky

0.79

ATIONAL

0.77

ewski

0.76

angles

0.74

ombie

0.74

ered

0.72

oppable

0.70

hips

0.70

agers

0.70

Activations Density 0.034%

terms related to authorization and organization activities

abbreviations or acronyms typically associated with organizations

No Comments

No Known Activations