INDEX

Explanations

mentions of the term "obs" at high activation levels

oai_token-act-pair · gpt-3.5-turbo

instances of the word "obs" and its variations in various contexts

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

âĸ¬

-0.79

 Sakuya

-0.71

ãĥĥãĥī

-0.63

 Karin

-0.62

 Poles

-0.60

 Monarch

-0.60

ART

-0.58

ä»£

-0.58

HER

-0.58

 Uriel

-0.57

POSITIVE LOGITS

essed

1.10

idian

1.05

poon

1.03

ession

1.00

ols

1.00

anding

0.99

essions

0.96

etting

0.95

poons

0.94

ourcing

0.93

Activations Density 0.008%

mentions of the term "obs" at high activation levels

instances of the word "obs" and its variations in various contexts

No Comments

No Known Activations