INDEX

Explanations

words or phrases related to arguments or debates

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

TRS

-0.15

chen

-0.15

akter

-0.15

igans

-0.15

zial

-0.15

 closure

-0.15

closure

-0.15

chsel

-0.14

ral

-0.14

eh

-0.14

POSITIVE LOGITS

uably

0.35

entin

0.32

onaut

0.31

entine

0.29

uing

0.28

entina

0.28

yle

0.27

uable

0.27

ued

0.27

inine

0.26

Activations Density 0.007%

words or phrases related to arguments or debates

No Comments

No Known Activations