INDEX

Explanations

phrases associated with job performance and results

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

-regexp

-0.15

ohl

-0.14

owie

-0.14

adil

-0.14

otron

-0.14

ad

-0.14

Ð¿Ð°

-0.14

ORIA

-0.14

angular

-0.13

undry

-0.13

POSITIVE LOGITS

cyan

0.15

kins

0.15

otten

0.15

enville

0.15

Md

0.15

verte

0.14

çľ

0.14

 squeeze

0.14

ieu

0.14

Anchor

0.14

Activations Density 0.002%

phrases associated with job performance and results

No Comments

No Known Activations