INDEX

Explanations

technical terms related to data analysis or technical acronyms

oai_token-act-pair · gpt-3.5-turbo

acronyms or terms related to specific organizations or systems

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

bidden

-0.75

liest

-0.75

urous

-0.74

aghan

-0.74

idity

-0.73

rette

-0.73

lier

-0.73

onyms

-0.72

ary

-0.71

arers

-0.71

POSITIVE LOGITS

ICS

1.51

FINE

0.93

INESS

0.93

FIR

0.88

YS

0.86

ECA

0.85

OURCE

0.85

 Flavoring

0.83

ENSE

0.83

OPA

0.83

Activations Density 0.016%

technical terms related to data analysis or technical acronyms

acronyms or terms related to specific organizations or systems

No Comments

No Known Activations