INDEX

Explanations

words related to strength, power, or authority

oai_token-act-pair · gpt-3.5-turbo

instances of the word "force" in various contexts

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

 Gleaming

-0.77

nia

-0.75

 Remastered

-0.75

çīĪ

-0.73

 weddings

-0.72

 Birthday

-0.66

 Wedding

-0.65

 Adventures

-0.64

anka

-0.63

 Columb

-0.63

POSITIVE LOGITS

 force

3.97

force

2.82

 Force

2.49

Force

2.07

 forces

2.00

forces

1.92

 Forces

1.68

 compel

1.41

forcing

1.34

 forcing

1.26

Activations Density 0.017%

words related to strength, power, or authority

instances of the word "force" in various contexts

No Comments

No Known Activations