INDEX

Explanations

the word 'only' followed by a number from 1 to 10

oai_token-act-pair · gpt-3.5-turbo

the word "only" or phrases indicating exclusivity

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

idon

-0.80

insula

-0.68

mass

-0.64

actionDate

-0.64

MAP

-0.63

align

-0.62

senal

-0.61

 Dynamics

-0.61

iths

-0.60

gnu

-0.60

POSITIVE LOGITS

 thing

0.82

 marginally

0.80

 drawback

0.77

 logical

0.74

 benef

0.73

 recourse

0.73

 valid

0.69

 half

0.69

 consolation

0.68

 surviving

0.68

Activations Density 0.045%

the word 'only' followed by a number from 1 to 10

the word "only" or phrases indicating exclusivity

No Comments

No Known Activations