INDEX

Explanations

words related to groups of people working together or collaborating, particularly in a professional or organized setting

oai_token-act-pair · gpt-3.5-turbo

references to groups or teams involved in various activities or contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

forward

-0.71

ously

-0.68

Film

-0.68

cot

-0.67

alam

-0.65

 Celeb

-0.62

ibrary

-0.62

 Franch

-0.62

CV

-0.60

icted

-0.59

POSITIVE LOGITS

members

1.31

 members

1.19

 mates

1.11

 member

1.09

member

1.02

peak

1.02

mates

0.98

sters

0.94

liquid

0.90

 Members

0.87

Activations Density 0.071%

words related to groups of people working together or collaborating, particularly in a professional or organized setting

references to groups or teams involved in various activities or contexts

No Comments

No Known Activations