INDEX

Explanations

adjectives related to responsiveness

oai_token-act-pair · gpt-3.5-turbo

terms related to responsiveness and attentiveness

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

obyl

-0.98

redit

-0.83

 Munich

-0.82

rug

-0.80

hammer

-0.79

ç¥ŀ

-0.77

Bav

-0.77

our

-0.73

################

-0.72

celona

-0.70

POSITIVE LOGITS

 responsive

1.06

 responsiveness

1.00

responsive

0.95

 intellig

0.83

ibilities

0.80

encies

0.80

ibly

0.76

 agre

0.71

 Respons

0.70

 element

0.70

Activations Density 0.014%

adjectives related to responsiveness

terms related to responsiveness and attentiveness

No Comments

No Known Activations