INDEX

Explanations

phrases related to achievement or advancement, particularly in professional or competitive contexts

oai_token-act-pair · gpt-3.5-turbo

terms related to significant events or positions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

terday

-0.67

 Yorkers

-0.64

atism

-0.61

enegger

-0.61

renheit

-0.60

utm

-0.60

sbm

-0.58

imon

-0.58

 sworn

-0.57

 Dear

-0.57

POSITIVE LOGITS

 fray

1.20

 throne

1.15

osphere

1.14

 realm

1.11

elight

1.06

 arena

1.05

 aisle

1.04

 forefront

1.03

 confines

0.99

 pinnacle

0.95

Activations Density 0.440%

phrases related to achievement or advancement, particularly in professional or competitive contexts

terms related to significant events or positions

No Comments

No Known Activations