INDEX

Explanations

words related to the term "Pod"

oai_token-act-pair · gpt-3.5-turbo

references to podcasts and their features

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

ãĤ¼ãĤ¦ãĤ¹

-0.84

tesy

-0.82

 Bermuda

-0.78

 Flavoring

-0.73

ãĥ¼ãĥ³

-0.71

 Samoa

-0.70

inventoryQuantity

-0.68

iveness

-0.68

ãĤ¤ãĥĪ

-0.67

 Colts

-0.65

POSITIVE LOGITS

Pod

1.31

casters

0.98

pod

0.97

caster

0.96

iatric

0.92

esta

0.91

Pod

0.89

ikarp

0.86

idon

0.82

emon

0.80

Activations Density 0.004%

words related to the term "Pod"

references to podcasts and their features

No Comments

No Known Activations