INDEX

Explanations

words related to infectious diseases, particularly Ebola and Zika

oai_token-act-pair · gpt-3.5-turbo

mentions of infectious diseases, particularly Ebola and Zika

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

FTWARE

-0.77

edom

-0.72

UGC

-0.71

UTC

-0.71

rawdownloadcloneembedreportprint

-0.71

 à¨

-0.71

iple

-0.70

kus

-0.70

*/(

-0.69

alties

-0.68

POSITIVE LOGITS

 virus

1.09

 outbreak

0.98

 toxin

0.98

 Ebola

0.96

 epidemic

0.94

 contag

0.94

 outbreaks

0.94

 Virus

0.93

 Outbreak

0.93

 infection

0.89

Activations Density 0.017%

words related to infectious diseases, particularly Ebola and Zika

mentions of infectious diseases, particularly Ebola and Zika

No Comments

No Known Activations