INDEX

Explanations

information related to a person's professional achievements and personal details

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

 Franti

-0.18

INVAL

-0.15

trinsic

-0.14

Ð»ÑĮÑĤ

-0.14

IGHL

-0.14

radu

-0.14

mpi

-0.14

ahoma

-0.14

/Instruction

-0.14

ngine

-0.14

POSITIVE LOGITS

 https

0.17

tog

0.16

hk

0.16

cheid

0.15

sb

0.15

Pod

0.15

0.14

COPY

0.14

Ur

0.14

arra

0.14

Activations Density 0.022%

information related to a person's professional achievements and personal details

No Comments

No Known Activations