INDEX

Explanations

nouns and phrases that denote concepts of condition, methodology, or relevance

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

 Lesser

-0.16

yles

-0.14

/org

-0.14

imli

-0.14

UDA

-0.14

iect

-0.14

inea

-0.13

inary

-0.13

ÑģÑĤÐ°

-0.13

 Sherman

-0.13

POSITIVE LOGITS

alysis

0.16

pto

0.16

 stead

0.15

empo

0.15

conds

0.14

å®ŀåľ¨

0.14

airo

0.14

ãĥĢãĤ¤

0.14

otope

0.14

orman

0.14

Activations Density 0.108%

nouns and phrases that denote concepts of condition, methodology, or relevance

No Comments

No Known Activations