INDEX

Explanations

words related to a specific object - "pods"

oai_token-act-pair · gpt-3.5-turbo

mentions of server pods and their associated counts

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

Pod

-0.81

 Chemistry

-0.72

ray

-0.72

rien

-0.69

Ren

-0.68

Der

-0.68

onne

-0.68

hern

-0.68

Fou

-0.67

 âĵĺ

-0.67

POSITIVE LOGITS

 pods

3.25

Kad

1.52

 capsules

1.45

pod

1.41

 clusters

1.14

Kiw

1.03

SAP

1.01

 capsule

1.01

 Pike

1.01

 Nordic

0.89

Activations Density 0.025%

words related to a specific object - "pods"

mentions of server pods and their associated counts

No Comments

No Known Activations