INDEX

Explanations

negative expressions related to disbelief or dissatisfaction

oai_token-act-pair · gpt-3.5-turbo

terms expressing a lack of faith or trust

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

anwhile

-0.86

 Madness

-0.77

uyomi

-0.75

piece

-0.73

å§«

-0.72

hyde

-0.72

 Shogun

-0.71

 Opera

-0.70

 Tile

-0.69

 Thornton

-0.68

POSITIVE LOGITS

athom

1.39

avour

1.29

riend

1.26

ortunately

1.23

ortun

1.17

ixed

1.11

ortunate

1.10

ruit

1.09

requently

1.08

requ

1.07

Activations Density 0.009%

negative expressions related to disbelief or dissatisfaction

terms expressing a lack of faith or trust

No Comments

No Known Activations