INDEX

Explanations

the word "willing", specifically in contexts where people are ready or agree to do something

oai_token-act-pair · gpt-3.5-turbo

expressions of willingness or commitment to actions or ideas

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

 Anthem

-0.84

hemy

-0.84

adish

-0.74

our

-0.73

rx

-0.69

VERTISEMENT

-0.69

allic

-0.65

arette

-0.65

ARS

-0.65

error

-0.65

POSITIVE LOGITS

theless

0.87

 enough

0.84

 gladly

0.77

terday

0.77

 willing

0.73

 unres

0.71

 willingly

0.69

 uncond

0.69

 bargain

0.67

 unwilling

0.65

Activations Density 0.029%

the word "willing", specifically in contexts where people are ready or agree to do something

expressions of willingness or commitment to actions or ideas

No Comments

No Known Activations