INDEX

Explanations

proper nouns or abbreviations

oai_token-act-pair · gpt-3.5-turbo

specific labels or identifiers related to organizations or regulatory bodies

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

aukee

-0.74

aminer

-0.72

redd

-0.71

antes

-0.71

mith

-0.69

 Dickinson

-0.68

 Whitman

-0.66

uddin

-0.66

 Twain

-0.65

MpServer

-0.65

POSITIVE LOGITS

PAL

1.50

acceptable

0.70

âķĲâķĲ

0.69

Ü

0.68

Official

0.67

The

0.67

SEA

0.66

 Official

0.66

ape

0.65

 Commission

0.65

Activations Density 0.018%

proper nouns or abbreviations

specific labels or identifiers related to organizations or regulatory bodies

No Comments

No Known Activations