INDEX

Explanations

references to active involvement or engagement in various contexts

oai_token-act-pair · gpt-3.5-turbo

instances of the word "participation."

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

 Damn

-0.74

oke

-0.67

 Detect

-0.63

pher

-0.63

Reviewed

-0.63

Clock

-0.61

leans

-0.61

 washed

-0.61

temp

-0.61

alled

-0.60

POSITIVE LOGITS

 participation

3.68

 Participation

2.60

particip

2.42

 involvement

2.08

Particip

1.95

 participating

1.74

 participate

1.74

 attendance

1.74

 Particip

1.71

 particip

1.69

Activations Density 0.007%

references to active involvement or engagement in various contexts

instances of the word "participation."

No Comments

No Known Activations