INDEX

Explanations

specific medical conditions, particularly hepatitis

oai_token-act-pair · gpt-3.5-turbo

terms related to medical conditions and personal health issues

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

•single token firings

fires strongly on " hepatitis", more weakly on "itis", "achelor", "assin", " epis", " Hep", " hepat".

No Comments

Negative Logits

erick

-1.00

eton

-0.92

ding

-0.82

*/(

-0.82

ed

-0.80

ded

-0.78

imental

-0.77

vich

-0.77

lar

-0.73

lain

-0.73

POSITIVE LOGITS

Ã©t

0.79

atform

0.75

atis

0.74

Wan

0.73

ques

0.71

xes

0.69

OE

0.67

omatic

0.67

 hepatitis

0.66

asca

0.66

Activations Density 0.101%

specific medical conditions, particularly hepatitis

terms related to medical conditions and personal health issues

No Comments

No Known Activations