INDEX

Explanations

phrases related to action or activity

oai_token-act-pair · gpt-3.5-turbo

phrases that emphasize action and activities

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

 herself

-0.84

olls

-0.71

agall

-0.70

estone

-0.69

mort

-0.68

 linen

-0.68

cients

-0.66

igans

-0.66

atten

-0.64

fol

-0.64

POSITIVE LOGITS

 inaction

0.82

ãĤŃ

0.73

 selector

0.73

psi

0.70

 mode

0.69

agonist

0.69

 Intent

0.69

 kinetic

0.68

 modes

0.68

Keefe

0.68

Activations Density 0.699%

phrases related to action or activity

phrases that emphasize action and activities

No Comments

No Known Activations