INDEX

Explanations

temporal references related to significant events or actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

[section

-0.15

aille

-0.14

GridColumn

-0.14

Kew

-0.14

obia

-0.13

trinsic

-0.13

aras

-0.12

RuntimeObject

-0.12

dic

-0.12

ãģ«ãģĭ

-0.12

POSITIVE LOGITS

 McMahon

0.15

croft

0.15

 interviewer

0.14

chio

0.14

enko

0.13

bÄĽ

0.13

bach

0.13

Ø²Ùĩ

0.13

cio

0.13

ocl

0.12

Activations Density 9.393%

temporal references related to significant events or actions

No Comments

No Known Activations