INDEX

Explanations

mentions of something being publicly announced or recognized as significant or important

oai_token-act-pair · gpt-3.5-turbo

terms related to significant announcements or proclamations

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

adra

-0.79

packing

-0.78

FO

-0.78

aho

-0.77

istence

-0.71

fts

-0.70

gger

-0.69

 drawer

-0.67

andro

-0.66

©¶æ

-0.65

POSITIVE LOGITS

 herald

1.06

ry

0.93

interstitial

0.90

eer

0.83

icons

0.77

 newcom

0.76

mares

0.75

ries

0.75

ed

0.72

ript

0.72

Activations Density 0.016%

mentions of something being publicly announced or recognized as significant or important

terms related to significant announcements or proclamations

No Comments

No Known Activations