INDEX

Explanations

terms related to awards or recognition

oai_token-act-pair · gpt-3.5-turbo

references to various awards

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

ths

-0.71

folk

-0.66

CY

-0.66

apse

-0.66

ullivan

-0.64

cos

-0.63

pher

-0.62

 breathing

-0.62

oidal

-0.62

wre

-0.61

POSITIVE LOGITS

 Awards

1.14

 Award

1.14

Winner

0.99

 Winners

0.96

 Prize

0.95

 nominations

0.95

 Winning

0.92

 Trophy

0.91

 awarded

0.88

 Fellowship

0.87

Activations Density 0.012%

terms related to awards or recognition

references to various awards

No Comments

No Known Activations