INDEX

Explanations

phrases related to actions or achievements

oai_token-act-pair · gpt-3.5-turbo

significant nouns related to actions, titles, and notable events

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

jri

-0.77

ifter

-0.76

iffe

-0.75

ovember

-0.75

thinkable

-0.74

ascript

-0.73

itta

-0.72

advertising

-0.72

amiya

-0.72

 compr

-0.71

POSITIVE LOGITS

 Monte

1.06

 Kirin

0.99

Uni

0.93

 Gran

0.92

Tri

0.92

 Raven

0.91

 Chrom

0.91

Gw

0.91

 Pulse

0.90

 Dunk

0.89

Activations Density 0.745%

phrases related to actions or achievements

significant nouns related to actions, titles, and notable events

No Comments

No Known Activations