INDEX

Explanations

abbreviations and acronyms commonly used in scientific or technical contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res_post_32k-oai

Configuration

jbloom/GPT2-Small-OAI-v5-32k-resid-post-SAEs/v5_32k_layer_10.pt

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

32,768

Data Type

torch.float32

Hook Name

blocks.10.hook_resid_post

Hook Layer

Architecture

standard

Context Size

Dataset

Skylion007/openwebtext

Activation Function

topk

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

lihood

-0.75

nian

-0.72

ende

-0.72

iques

-0.72

nir

-0.66

 attent

-0.65

ness

-0.64

lers

-0.64

lings

-0.63

Pwr

-0.61

POSITIVE LOGITS

VE

1.30

UGH

1.25

uthor

1.19

BILITY

1.18

BLE

1.13

IL

1.13

KE

1.13

ccess

1.11

UTH

1.10

ILS

1.09

Activations Density 0.207%

abbreviations and acronyms commonly used in scientific or technical contexts

No Comments

No Known Activations