INDEX

Explanations

phrases related to political terms and government structures

oai_token-act-pair · gpt-3.5-turbo

phrases that express a hierarchical relationship or ranking among groups or individuals

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

mith

-0.86

ozo

-0.78

hirt

-0.72

hift

-0.71

HT

-0.69

nw

-0.67

enegger

-0.66

ascript

-0.66

-0.65

-0.63

POSITIVE LOGITS

 parcel

0.86

 file

0.80

file

0.77

 dues

0.69

 accomplishment

0.69

idates

0.65

 gown

0.63

 conditioning

0.62

 testament

0.62

rity

0.61

Activations Density 0.072%

phrases related to political terms and government structures

phrases that express a hierarchical relationship or ranking among groups or individuals

No Comments

No Known Activations