INDEX

Explanations

words related to earning or receiving rewards and recognition

oai_token-act-pair · gpt-3.5-turbo

the concept of "ear" in various contexts

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Tess

-0.66

 Cuomo

-0.66

 Springfield

-0.66

VK

-0.66

 Grav

-0.65

 Athena

-0.63

 Churchill

-0.60

 Punk

-0.59

ISS

-0.59

Ts

-0.59

POSITIVE LOGITS

ear

1.34

earing

1.07

nings

1.07

ears

0.99

lier

0.97

unci

0.90

ning

0.88

liest

0.85

ably

0.85

gypt

0.84

Activations Density 0.004%

words related to earning or receiving rewards and recognition

the concept of "ear" in various contexts

No Comments

No Known Activations