INDEX

Explanations

names of political figures

oai_token-act-pair · gpt-3.5-turbo

individual letters or single-character variables

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

Measure

-0.72

 convol

-0.67

 subtract

-0.64

 dispatcher

-0.64

 feder

-0.63

 cryst

-0.63

CHO

-0.62

editor

-0.60

 depress

-0.59

ept

-0.59

POSITIVE LOGITS

uala

0.94

shi

0.93

icz

0.85

ian

0.84

ius

0.80

ÃŃa

0.79

ny

0.77

itic

0.77

itsch

0.77

nikov

0.75

Activations Density 0.186%

names of political figures

individual letters or single-character variables

No Comments

No Known Activations