INDEX

Explanations

dates and numerical information

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 15-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L15R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.15.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

ruc

-0.16

atcher

-0.16

 juris

-0.14

halt

-0.14

aman

-0.14

Jur

-0.14

accum

-0.14

enumerator

-0.14

wolf

-0.14

luck

-0.13

POSITIVE LOGITS

th

0.24

rd

0.17

LAG

0.17

st

0.16

0.15

Ð³Ð°Ð½

0.15

fly

0.15

aines

0.14

Lag

0.14

ÑĥÐ²Ð°

0.14

Activations Density 0.181%

dates and numerical information

No Comments

No Known Activations