INDEX

Explanations

phrases related to calls for actions or changes in public policy

oai_token-act-pair · gpt-3.5-turbo

calls for action, demands, or changes 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 @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 coached

-0.68

hops

-0.62

 biod

-0.62

ths

-0.60

 Moons

-0.59

ooz

-0.58

hog

-0.57

ago

-0.57

 perfectly

-0.57

 fart

-0.57

POSITIVE LOGITS

 reconsider

0.88

venge

0.83

 urgently

0.81

erity

0.76

amaz

0.73

ãĤ¹

0.72

 attention

0.71

 urgent

0.71

 clarification

0.68

 forgiveness

0.68

Activations Density 0.391%

phrases related to calls for actions or changes in public policy

calls for action, demands, or changes in various contexts

No Comments

No Known Activations