INDEX

Explanations

words or acronyms related to transportation or measurement units

oai_token-act-pair · gpt-3.5-turbo

references to specific research articles or academic formatting

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

 convol

-0.72

BIP

-0.71

 exha

-0.68

 incent

-0.68

 probation

-0.68

 reluct

-0.67

merce

-0.67

 bargaining

-0.65

 Bind

-0.64

 predec

-0.62

POSITIVE LOGITS

ional

1.01

ory

0.95

omatic

0.90

olkien

0.90

iary

0.89

iger

0.86

anu

0.86

acker

0.85

eless

0.85

icism

0.85

Activations Density 0.013%

words or acronyms related to transportation or measurement units

references to specific research articles or academic formatting

No Comments

No Known Activations