INDEX

Explanations

references to personal awareness or knowledge

oai_token-act-pair · gpt-3.5-turbo

first-person pronouns and expressions of awareness or capability

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

 Boat

-0.62

quished

-0.59

 Travels

-0.58

itudes

-0.58

TIME

-0.58

 Imper

-0.57

 Institution

-0.57

rus

-0.57

 Rout

-0.56

 Nguyen

-0.56

POSITIVE LOGITS

 concerned

1.12

 know

0.86

 ascertain

0.86

 aware

0.82

 remember

0.81

 recall

0.81

 remem

0.80

 knowledge

0.77

knowledge

0.76

 recollection

0.75

Activations Density 0.068%

references to personal awareness or knowledge

first-person pronouns and expressions of awareness or capability

No Comments

No Known Activations