INDEX

Explanations

verbs indicating involvement in an activity or event

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "participate" and its variations in the context of involvement

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

forward

-0.66

esome

-0.66

riz

-0.62

borne

-0.62

pe

-0.61

pex

-0.60

erity

-0.59

flies

-0.59

dylib

-0.56

Nau

-0.52

POSITIVE LOGITS

in

0.92

 therein

0.76

 actively

0.73

atta

0.73

 academ

0.69

 Participation

0.68

 guiActiveUn

0.68

 prominently

0.67

 participate

0.64

 Particip

0.62

Activations Density 0.059%

verbs indicating involvement in an activity or event

occurrences of the word "participate" and its variations in the context of involvement

No Comments

No Known Activations