INDEX

Explanations

phrases related to various actions and events

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "actions" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

wn

-0.70

lund

-0.66

 Saban

-0.65

ille

-0.64

lot

-0.64

KER

-0.64

onite

-0.63

zzo

-0.62

rette

-0.62

 topped

-0.62

POSITIVE LOGITS

 actions

3.64

 Actions

2.68

 action

2.25

 ACTIONS

2.22

 deeds

2.04

actions

1.83

 activities

1.83

 behaviors

1.81

 inaction

1.73

 behaviours

1.68

Activations Density 0.014%

phrases related to various actions and events

occurrences of the word "actions" in various contexts

No Comments

No Known Activations