INDEX

Explanations

phrases related to completeness or entirety

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "whole."

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

nance

-0.68

aru

-0.63

ime

-0.63

ciplinary

-0.62

 Advis

-0.60

onse

-0.58

arius

-0.57

 aboard

-0.57

istics

-0.56

ãĤ¨

-0.55

POSITIVE LOGITS

 whole

3.73

 entire

2.68

 entirety

1.92

 Whole

1.84

ENT

1.25

 rest

1.05

 totality

1.04

 full

1.03

 entirely

1.00

 totally

0.98

Activations Density 0.015%

phrases related to completeness or entirety

occurrences of the word "whole."

No Comments

No Known Activations