INDEX

Explanations

phrases related to collaboration and unity

oai_token-act-pair · gpt-3.5-turbo

instances of the word "together."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

advant

-0.81

occ

-0.72

dark

-0.67

articles

-0.67

ble

-0.64

cock

-0.64

odan

-0.64

null

-0.62

ck

-0.62

anty

-0.62

POSITIVE LOGITS

 arrang

0.73

 affili

0.72

ieth

0.71

halla

0.71

essions

0.71

displayText

0.71

essional

0.70

 endeav

0.69

ãĤ¤ãĥĪ

0.67

 bargaining

0.66

Activations Density 0.021%

phrases related to collaboration and unity

instances of the word "together."

No Comments

No Known Activations