INDEX

Explanations

locations around the world

oai_token-act-pair · gpt-3.5-turbo

references to specific individuals, entities, or places

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

ãĥ¼ãĥĨ

-0.68

 lean

-0.64

HQ

-0.63

unity

-0.62

furt

-0.61

rency

-0.59

cffffcc

-0.59

uese

-0.58

psi

-0.57

 surrog

-0.57

POSITIVE LOGITS

ovych

0.74

ulic

0.73

horn

0.73

cock

0.71

hoff

0.64

glers

0.62

oshenko

0.61

lich

0.61

eps

0.61

 Wand

0.61

Activations Density 0.728%

locations around the world

references to specific individuals, entities, or places

No Comments

No Known Activations