INDEX

Explanations

mentions of a specific person's name, "Koh"

oai_token-act-pair · gpt-3.5-turbo

proper nouns, specifically names

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

xual

-0.89

*/(

-0.73

phia

-0.71

PsyNetMessage

-0.70

 balloons

-0.69

opausal

-0.69

conserv

-0.66

éĹĺ

-0.66

 Canadiens

-0.66

ngth

-0.64

POSITIVE LOGITS

lyak

0.98

istani

0.88

aku

0.84

len

0.84

sten

0.82

ota

0.82

kus

0.81

Koh

0.81

eling

0.78

adish

0.78

Activations Density 0.017%

mentions of a specific person's name, "Koh"

proper nouns, specifically names

No Comments

No Known Activations