INDEX

Explanations

names or entities containing the word "Jen"

oai_token-act-pair · gpt-3.5-turbo

references to individuals named Jen or Jenkins

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

ricular

-0.74

tin

-0.69

Interstitial

-0.69

 Portug

-0.68

yright

-0.64

tti

-0.64

ties

-0.64

gres

-0.63

anguages

-0.63

beit

-0.63

POSITIVE LOGITS

esis

1.02

eter

0.80

etts

0.74

jen

0.74

eton

0.72

icter

0.71

chell

0.71

 Jenkins

0.71

quez

0.68

ova

0.68

Activations Density 0.078%

names or entities containing the word "Jen"

references to individuals named Jen or Jenkins

No Comments

No Known Activations