INDEX

Explanations

terms related to a selection process or meeting specific requirements

oai_token-act-pair · gpt-3.5-turbo

terms related to qualifying events or criteria

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

 tentacles

-0.70

aneers

-0.68

 Statue

-0.67

antine

-0.66

 Troll

-0.65

MpServer

-0.65

Sew

-0.63

anse

-0.62

Scale

-0.62

nc

-0.62

POSITIVE LOGITS

 qualifying

3.77

 qualifiers

2.77

 qualifier

2.66

 Qual

2.61

 qualification

2.36

Qual

2.34

 qualify

2.12

 qualified

1.95

qual

1.93

 qualifies

1.71

Activations Density 0.025%

terms related to a selection process or meeting specific requirements

terms related to qualifying events or criteria

No Comments

No Known Activations