INDEX

Explanations

mentions of numbers in various contexts

oai_token-act-pair · gpt-3.5-turbo

references to decision-making processes or choices

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

.ãĢį

-0.66

.</

-0.66

[/

-0.63

''.

-0.61

âĢ¦âĢ¦

-0.58

âĢ¦."

-0.57

âĢ¦..

-0.57

!".

-0.56

.—

-0.56

âĶĢâĶĢâĶĢâĶĢâĶĢâĶĢâĶĢâĶĢ

-0.56

POSITIVE LOGITS

 different

0.43

 similarities

0.41

 similarity

0.40

 differing

0.39

 difference

0.38

 differences

0.36

different

0.36

 pairs

0.35

 divide

0.34

esides

0.34

Activations Density 2.302%

mentions of numbers in various contexts

references to decision-making processes or choices

No Comments

No Known Activations