INDEX

Explanations

locations or places as indicated by the word "home"

oai_token-act-pair · gpt-3.5-turbo

references to places being described as "home."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

entle

-0.58

 Annotations

-0.56

 Spir

-0.55

 Canter

-0.54

 Seym

-0.53

 Genie

-0.53

geries

-0.53

 asymm

-0.52

ucket

-0.51

 Sense

-0.51

POSITIVE LOGITS

coming

0.87

liest

0.84

opathy

0.83

ported

0.79

amed

0.78

opathic

0.78

grown

0.77

brew

0.77

pages

0.72

cffffcc

0.72

Activations Density 0.029%

locations or places as indicated by the word "home"

references to places being described as "home."

No Comments

No Known Activations