INDEX

Explanations

mentions of political figures and elections

oai_token-act-pair · gpt-3.5-turbo

phrases concerning majority opinions or statistics

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

gage

-0.77

 surv

-0.72

 accomp

-0.70

 crus

-0.68

wu

-0.68

wyn

-0.67

arily

-0.66

 unden

-0.66

 distingu

-0.65

ndra

-0.65

POSITIVE LOGITS

 Helpful

0.72

Sof

0.67

 Hurt

0.66

Parameters

0.65

 Hydro

0.64

 Kund

0.64

iston

0.63

 Hansen

0.63

 Dumb

0.62

 Lank

0.62

Activations Density 0.000%

mentions of political figures and elections

phrases concerning majority opinions or statistics

No Comments

No Known Activations