INDEX

Explanations

symbols or formatting markers indicating structural divisions or categories within text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 6-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L6R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.6.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

uffman

-0.16

Ð¾Ð¿Ð¾Ð»

-0.15

earer

-0.15

çļĦæīĭ

-0.15

uhn

-0.15

@nate

-0.15

ynch

-0.15

ught

-0.14

icum

-0.14

enberg

-0.14

POSITIVE LOGITS

 Lind

0.15

 Lands

0.14

èŀº

0.14

Ãĭ

0.14

 Erik

0.14

«

0.14

inary

0.13

 surf

0.13

Rap

0.13

oints

0.13

Activations Density 0.017%

symbols or formatting markers indicating structural divisions or categories within text

No Comments

No Known Activations