INDEX

Explanations

phrases related to actions that involve appearing or arriving in a particular situation or place

oai_token-act-pair · gpt-3.5-turbo

instances of the word "came."

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

apixel

-0.91

²¾

-0.89

olor

-0.87

icons

-0.81

illusion

-0.81

sav

-0.80

heed

-0.79

amaz

-0.79

lav

-0.78

raid

-0.78

POSITIVE LOGITS

 undone

1.06

 ashore

0.82

 WRITE

0.79

out

0.73

OUT

0.72

 forth

0.72

 Gerrard

0.71

 Lans

0.69

Pes

0.68

BuyableInstoreAndOnline

0.68

Activations Density 0.027%

phrases related to actions that involve appearing or arriving in a particular situation or place

instances of the word "came."

No Comments

No Known Activations