INDEX

Explanations

mentions of energy upgrades related to a local program

oai_token-act-pair · gpt-3.5-turbo

instances of asterisks or stars used as markers or highlights

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

 spir

-0.69

 manifold

-0.68

oidal

-0.67

 scattering

-0.66

 erosion

-0.65

 absorbed

-0.65

cephal

-0.65

acle

-0.65

 neighb

-0.64

zza

-0.62

POSITIVE LOGITS

WARNING

1.09

THIS

1.06

UPDATE

1.01

DOWN

0.95

NEW

0.94

EDIT

0.94

NOTE

0.94

Update

0.93

Discussion

0.90

!/

0.90

Activations Density 0.022%

mentions of energy upgrades related to a local program

instances of asterisks or stars used as markers or highlights

No Comments

No Known Activations