INDEX

Explanations

references to time stamps and digital identifiers

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 10-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L10R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.10.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

amt

-0.15

atos

-0.15

ako

-0.15

Hud

-0.14

ag

-0.14

amm

-0.14

erg

-0.14

iale

-0.14

add

-0.13

pan

-0.13

POSITIVE LOGITS

issan

0.17

inia

0.15

 Cumberland

0.15

 interval

0.14

//===

0.14

removeAttr

0.14

æģ¯

0.14

orst

0.14

zers

0.14

Activations Density 0.015%

references to time stamps and digital identifiers

No Comments

No Known Activations