INDEX

Explanations

references to specific entities or products, likely related to reviews or recommendations

oai_token-act-pair · gpt-3.5-turbo

occurrences of the letter 's'

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

pressed

-0.64

requisites

-0.64

 hitch

-0.63

cfg

-0.63

zos

-0.63

ainers

-0.62

bp

-0.62

ilater

-0.60

batch

-0.59

ISON

-0.59

POSITIVE LOGITS

 Revenge

1.03

 newest

0.98

kaya

0.97

ullivan

0.88

 finest

0.83

 Choice

0.81

ELF

0.80

ï¸

0.78

niper

0.78

 Greatest

0.77

Activations Density 0.180%

references to specific entities or products, likely related to reviews or recommendations

occurrences of the letter 's'

No Comments

No Known Activations