INDEX

Explanations

phrases that indicate significant contributions or achievements in a professional context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

ÐµÐºÑĤÐ¸

-0.15

retched

-0.14

entiful

-0.14

 ãĥ¯

-0.14

iba

-0.14

Ð»Ð¸Ð·

-0.14

æ±¡

-0.14

itched

-0.14

ocking

-0.14

ÑĦÐ¾ÑĢÐ¼Ð¸

-0.14

POSITIVE LOGITS

ano

0.17

igel

0.15

 loos

0.14

alian

0.13

hol

0.13

.scalablytyped

0.13

apgolly

0.13

 Curtain

0.13

 Jude

0.13

Ð´Ð¾Ð½

0.13

Activations Density 0.060%

phrases that indicate significant contributions or achievements in a professional context

No Comments

No Known Activations