INDEX

Explanations

phrases related to willingness or openness

oai_token-act-pair · gpt-3.5-turbo

expressions of willingness or intent to take action

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

 Compass

-0.65

 Sina

-0.65

Statistics

-0.64

availability

-0.62

iens

-0.62

Bound

-0.61

calling

-0.60

ndra

-0.60

ants

-0.60

Textures

-0.59

POSITIVE LOGITS

 sacrifice

1.28

 concede

1.27

 tolerate

1.22

 compromise

1.17

 accept

1.14

 gamble

1.13

 overlook

1.12

 cooperate

1.12

 forgive

1.04

 sacrific

1.03

Activations Density 0.108%

phrases related to willingness or openness

expressions of willingness or intent to take action

No Comments

No Known Activations