INDEX

Explanations

URLs and links to digital content

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

ERAL

-0.15

illac

-0.14

ijd

-0.14

:CGRect

-0.13

iegel

-0.13

 Heal

-0.13

tru

-0.13

 onDataChange

-0.13

elijk

-0.13

OUSE

-0.13

POSITIVE LOGITS

aty

0.16

aland

0.15

ret

0.15

 inferred

0.15

hit

0.14

abinet

0.14

fur

0.14

phies

0.14

Pep

0.14

Activations Density 0.018%

URLs and links to digital content

No Comments

No Known Activations