INDEX

Explanations

phrases indicating examples or samples of various items or concepts

oai_token-act-pair · gpt-3.5-turbo

instances of the word "such" followed by explanations or examples

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

ertodd

-0.76

zl

-0.72

olate

-0.71

itudinal

-0.71

ipedia

-0.70

rition

-0.67

oil

-0.67

hene

-0.66

 Drum

-0.66

kick

-0.66

POSITIVE LOGITS

ties

0.78

 minded

0.68

 consequential

0.66

cond

0.66

should

0.62

 aggreg

0.60

 abundantly

0.60

things

0.60

minded

0.59

 constituted

0.59

Activations Density 0.050%

phrases indicating examples or samples of various items or concepts

instances of the word "such" followed by explanations or examples

No Comments

No Known Activations