INDEX

Explanations

instances of the word "active."

oai_token-act-pair · gpt-3.5-turbo

instances of the word "active" 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 @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

OUGH

-0.78

elf

-0.73

unts

-0.69

 Grind

-0.69

umbo

-0.68

aneers

-0.68

awed

-0.67

Luk

-0.66

adra

-0.66

atered

-0.66

POSITIVE LOGITS

 active

1.17

 inactive

0.94

 newsp

0.89

active

0.88

 participation

0.85

 ingredient

0.78

activity

0.78

 volcano

0.78

 involvement

0.77

 Active

0.77

Activations Density 0.008%

instances of the word "active."

instances of the word "active" in various contexts

No Comments

No Known Activations