INDEX

Explanations

sentences describing things that are in a bad or unfavorable condition

oai_token-act-pair · gpt-3.5-turbo

instances of the word "badly" used in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

anwhile

-0.84

orney

-0.77

iferation

-0.70

recy

-0.69

ithe

-0.68

utral

-0.68

azi

-0.68

igg

-0.67

hip

-0.66

iry

-0.66

POSITIVE LOGITS

 badly

0.96

 beaten

0.89

 enough

0.84

 adolesc

0.81

 patched

0.76

 redes

0.74

WARE

0.73

 damaged

0.73

 inflicted

0.71

 suited

0.70

Activations Density 0.003%

sentences describing things that are in a bad or unfavorable condition

instances of the word "badly" used in various contexts

No Comments

No Known Activations