INDEX

Explanations

specific phrases and terminologies related to instructions and guidance within a context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

opsis

-0.15

iegel

-0.14

éģ£

-0.13

ics

-0.13

erson

-0.13

Æ¡i

-0.13

Ð²Ð¾Ð»

-0.13

heim

-0.13

 váºŃt

-0.13

POSITIVE LOGITS

[[]

0.17

å©Ĩ

0.15

 .↵

0.15

ãĢħ

0.14

oÄŁ

0.13

ensi

0.13

 Undefined

0.13

inke

0.13

 Unknown

0.13

Î»Î¿Î³

0.13

Activations Density 0.200%

specific phrases and terminologies related to instructions and guidance within a context

No Comments

No Known Activations