INDEX

Explanations

phrases related to taking action or activism

oai_token-act-pair · gpt-3.5-turbo

references to specific calls to action or initiatives

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

 plateau

-0.73

lf

-0.65

 cottage

-0.65

 conce

-0.65

¾

-0.63

 eclips

-0.63

ettel

-0.63

wn

-0.61

ld

-0.61

ome

-0.61

POSITIVE LOGITS

 Action

3.90

Action

2.79

 ACTION

2.63

 action

2.41

 Actions

2.10

action

1.98

 ACTIONS

1.42

 actions

1.38

ACTION

1.38

actions

1.23

Activations Density 0.013%

phrases related to taking action or activism

references to specific calls to action or initiatives

No Comments

No Known Activations