INDEX

Explanations

mentions of terms related to medical issues and mortality rates

oai_token-act-pair · gpt-3.5-turbo

references to mortality rates and related concepts

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

kef

-0.99

ledge

-0.81

ube

-0.75

wcsstore

-0.74

anooga

-0.72

KER

-0.72

afort

-0.72

arte

-0.71

venants

-0.70

rupulous

-0.70

POSITIVE LOGITS

 mortality

0.97

 contag

0.96

 expectancy

0.94

 regress

0.88

 sickness

0.85

 remission

0.78

 prevention

0.77

ously

0.76

 complications

0.75

 psychiat

0.75

Activations Density 0.027%

mentions of terms related to medical issues and mortality rates

references to mortality rates and related concepts

No Comments

No Known Activations