INDEX

Explanations

terms related to strategic planning or initiatives

oai_token-act-pair · gpt-3.5-turbo

references to strategic concepts or terminology

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

 bend

-0.85

 torment

-0.80

 pity

-0.75

 oppress

-0.74

 compat

-0.73

fil

-0.71

 maid

-0.70

 muff

-0.70

 indifference

-0.69

ap

-0.69

POSITIVE LOGITS

 Strategic

3.30

ategic

1.73

 Strategy

1.39

 Tactical

1.37

 Comprehensive

1.32

 Strategies

1.24

Str

1.18

STR

1.17

 Aerospace

1.16

 Strateg

1.14

Activations Density 0.053%

terms related to strategic planning or initiatives

references to strategic concepts or terminology

No Comments

No Known Activations