INDEX

Explanations

keywords related to the potential uses or applications of technology or objects

oai_token-act-pair · gpt-3.5-turbo

references to various applications or uses of a technology

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Bagg

-0.62

 Tale

-0.61

 Arist

-0.60

 Panda

-0.59

AAA

-0.59

 Pill

-0.58

 Polly

-0.58

 Gleaming

-0.57

HB

-0.57

 nerv

-0.56

POSITIVE LOGITS

fully

0.82

 applications

0.78

ability

0.74

cases

0.74

benefit

0.72

fulness

0.72

entials

0.70

iences

0.70

etimes

0.70

ivo

0.69

Activations Density 0.101%

keywords related to the potential uses or applications of technology or objects

references to various applications or uses of a technology

No Comments

No Known Activations