INDEX

Explanations

mathematical expressions and symbols

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

nech

-0.17

tah

-0.15

itto

-0.15

viÄį

-0.15

alis

-0.15

acific

-0.14

ÏĢÎ¬

-0.14

aja

-0.14

ought

-0.13

 shares

-0.13

POSITIVE LOGITS

0.17

avel

0.15

Poe

0.15

ottom

0.15

 flood

0.14

erate

0.14

weise

0.14

ulpt

0.13

ÏħÎº

0.13

Activations Density 0.030%

mathematical expressions and symbols

No Comments

No Known Activations