INDEX

Explanations

references to academic citations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

ê±´

-0.15

IGHL

-0.14

rial

-0.14

unte

-0.14

ambia

-0.14

stantiate

-0.13

ohan

-0.13

entine

-0.13

assi

-0.13

rowsers

-0.13

POSITIVE LOGITS

onis

0.15

aina

0.14

vfs

0.14

ailer

0.14

cf

0.14

ubyte

0.13

see

0.13

uly

0.13

 Ferr

0.13

Activations Density 0.009%

references to academic citations

No Comments

No Known Activations