INDEX

Explanations

specific formatting rules or information in a structured data format

oai_token-act-pair · gpt-3.5-turbo

numerical data or statistics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

 subsequ

-0.82

 servicing

-0.73

 rule

-0.69

 fres

-0.67

 tremend

-0.67

aptic

-0.67

bered

-0.67

 eaves

-0.66

reon

-0.66

 forwarding

-0.66

POSITIVE LOGITS

 None

1.53

TBD

1.41

 Unknown

1.38

TBA

1.33

 Various

1.24

 Provided

1.12

 Played

1.10

 Approximately

1.10

???

1.09

 Available

1.08

Activations Density 0.090%

specific formatting rules or information in a structured data format

numerical data or statistics

No Comments

No Known Activations