INDEX

Explanations

technical terms and error codes related to software and APIs

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 7-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L7R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.7.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

ollen

-0.15

ansson

-0.15

avar

-0.14

ivet

-0.14

 Sylvia

-0.14

ingly

-0.14

rahim

-0.14

ikes

-0.14

subs

-0.14

 ceremon

-0.13

POSITIVE LOGITS

ozo

0.17

achie

0.15

agli

0.14

 Daly

0.14

<strong

0.14

 Verd

0.14

çīĮ

0.13

meg

0.13

Fir

0.13

ovaly

0.13

Activations Density 0.001%

technical terms and error codes related to software and APIs

No Comments

No Known Activations