INDEX

Explanations

references to unique identifiers or characteristics in scientific research

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 13-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L13R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.13.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

ubat

-0.16

ttp

-0.15

Ãłng

-0.15

iena

-0.15

@Table

-0.15

ToJson

-0.14

ilon

-0.14

.der

-0.14

 pupil

-0.14

drv

-0.13

POSITIVE LOGITS

cha

0.19

 shame

0.15

aid

0.15

edList

0.14

-t

0.14

clud

0.14

halten

0.14

hta

0.14

mrt

0.14

Activations Density 0.001%

references to unique identifiers or characteristics in scientific research

No Comments

No Known Activations