INDEX

Explanations

references to professional or academic societies

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

nk

-0.15

andles

-0.15

 Tone

-0.15

OOSE

-0.15

cow

-0.15

uese

-0.14

.TableName

-0.14

zen

-0.14

ãĤīãģĽ

-0.14

emmel

-0.14

POSITIVE LOGITS

apers

0.16

achuset

0.15

uite

0.15

NDER

0.14

illum

0.14

artz

0.14

 Certified

0.14

CJK

0.13

stra

0.13

egl

0.13

Activations Density 0.007%

references to professional or academic societies

No Comments

No Known Activations