INDEX

Explanations

negative descriptive terms

oai_token-act-pair · gpt-3.5-turbo

strong evaluations of situations or conditions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

 Winn

-0.76

ien

-0.72

ogi

-0.70

nos

-0.70

oday

-0.69

 Woodward

-0.69

 Hodg

-0.68

 introdu

-0.67

Sok

-0.67

 Shepherd

-0.65

POSITIVE LOGITS

etary

0.79

 pixel

0.74

bara

0.72

;;

0.71

android

0.71

dat

0.71

gow

0.71

ãĥĦ

0.69

FLAG

0.69

Ĥ

0.69

Activations Density 0.451%

negative descriptive terms

strong evaluations of situations or conditions

No Comments

No Known Activations