INDEX

Explanations

key terms related to the impact of various actions, events, or policies in different contexts

oai_token-act-pair · gpt-3.5-turbo

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

phabet

-0.78

 Pages

-0.56

 pleading

-0.54

 condol

-0.54

ciation

-0.54

 begg

-0.53

 tracks

-0.53

lig

-0.53

Dial

-0.52

 wondering

-0.52

POSITIVE LOGITS

¶ħ

0.87

Had

0.71

 exert

0.69

arella

0.69

senal

0.66

ãĥķãĤ©

0.66

 exerted

0.66

entially

0.66

Thumbnail

0.65

ãĥŁ

0.65

Activations Density 0.115%

key terms related to the impact of various actions, events, or policies in different contexts

No Comments

No Known Activations