INDEX

Explanations

names of academic institutions or locations within contexts

oai_token-act-pair · gpt-3.5-turbo

proper nouns, specifically names of people and places

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ocol

-0.74

è¦ļéĨĴ

-0.73

 indo

-0.70

wana

-0.64

 behav

-0.63

 Nemesis

-0.63

 Allies

-0.63

obia

-0.62

 behavi

-0.60

 affili

-0.60

POSITIVE LOGITS

ayette

0.97

lean

0.87

hyde

0.87

ville

0.85

veland

0.84

prise

0.84

stown

0.82

ulton

0.81

>>>>>>>>

0.79

etics

0.79

Activations Density 0.032%

names of academic institutions or locations within contexts

proper nouns, specifically names of people and places

No Comments

No Known Activations