INDEX

Explanations

phrases related to positions or roles within a team or organization

oai_token-act-pair · gpt-3.5-turbo

elements related to production roles and team collaborations in film or media

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

nesty

-0.67

 millionaires

-0.66

itiveness

-0.65

 arous

-0.62

fw

-0.62

Downloadha

-0.62

yip

-0.61

hematically

-0.61

illions

-0.60

urers

-0.59

POSITIVE LOGITS

 helm

1.20

 vocals

1.01

 duties

0.97

 interim

0.90

 narration

0.90

lead

0.88

 reins

0.87

 narr

0.86

 overseeing

0.85

 directing

0.85

Activations Density 0.278%

phrases related to positions or roles within a team or organization

elements related to production roles and team collaborations in film or media

No Comments

No Known Activations