INDEX

Explanations

references to titles, entities, and significant political or economic terms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

ardon

-0.16

uing

-0.16

antine

-0.15

amble

-0.15

©

-0.15

abal

-0.15

lik

-0.14

itting

-0.14

awi

-0.14

uster

-0.14

POSITIVE LOGITS

ãĥ»ãĥ»ãĥ»↵↵

0.14

UME

0.14

_charset

0.14

zer

0.14

JNIEnv

0.14

ê´Ģ

0.14

 Cumhur

0.14

aux

0.14

 sola

0.14

enza

0.13

Activations Density 0.007%

references to titles, entities, and significant political or economic terms

No Comments

No Known Activations

references to titles, entities, and significant political or economic terms

No Comments

No Known Activations