INDEX

Explanations

terms related to classification, definition, and description of characteristics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

emos

-0.17

oga

-0.16

erca

-0.15

enÃŃ

-0.14

adr

-0.14

ÅĽÄĩ

-0.14

ÐµÑĤÑģÑı

-0.14

ER

-0.14

abant

-0.14

eree

-0.14

POSITIVE LOGITS

ire

0.35

ir

0.28

irc

0.25

irl

0.25

irm

0.25

ite

0.24

isci

0.23

isce

0.23

idor

0.23

IRE

0.23

Activations Density 0.028%

terms related to classification, definition, and description of characteristics

No Comments

No Known Activations