INDEX

Explanations

mentions of the name "Hobs" with varying intensities

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "obs."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

ion

-0.68

 transmitter

-0.68

 nurse

-0.67

 recorded

-0.64

 afternoon

-0.61

 intercept

-0.61

 medicine

-0.59

 Vand

-0.59

 vector

-0.58

Ell

-0.58

POSITIVE LOGITS

obs

4.77

ob

1.71

obb

1.37

obby

1.26

OB

1.22

obbies

1.22

oby

1.18

oba

1.09

azes

1.04

abs

1.02

Activations Density 0.009%

mentions of the name "Hobs" with varying intensities

occurrences of the word "obs."

No Comments

No Known Activations