INDEX

Explanations

Information related to the word "color."

oai_token-act-pair · gpt-3.5-turbo

references to color

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

Way

-0.74

ARK

-0.66

 Malaysian

-0.63

 Danish

-0.63

 Manifest

-0.61

 Norwegian

-0.61

Phase

-0.60

 Gingrich

-0.60

 grounding

-0.59

ELF

-0.59

POSITIVE LOGITS

ossus

1.34

iseum

1.22

ophon

1.14

onel

1.12

umn

1.00

orsche

1.00

lections

0.99

oqu

0.99

ossal

0.98

estial

0.97

Activations Density 0.020%

Information related to the word "color."

references to color

No Comments

No Known Activations