INDEX

Explanations

words related to inspiration or influence

oai_token-act-pair · gpt-3.5-turbo

references to inspiration and involvement 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 @ 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

Zone

-0.60

 mouths

-0.57

aptic

-0.57

 ILCS

-0.57

 whats

-0.56

ugs

-0.56

"},{"

-0.55

="/

-0.54

 coasts

-0.51

 mortals

-0.51

POSITIVE LOGITS

 earlier

1.13

 last

1.08

 yesterday

0.98

 initially

0.95

 beforehand

0.90

 recently

0.89

 unsuccessfully

0.89

 briefly

0.87

 during

0.85

 originally

0.84

Activations Density 0.345%

words related to inspiration or influence

references to inspiration and involvement in various contexts

No Comments

No Known Activations