INDEX

Explanations

elements related to publications and their editorial staff

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Curtis

-0.15

quier

-0.15

aic

-0.15

icine

-0.15

 Criterion

-0.15

iÄį

-0.14

apas

-0.14

CTS

-0.14

 Dictionary

-0.14

ãĥ¼ãĤ

-0.13

POSITIVE LOGITS

afort

0.16

å°¾

0.14

ITAL

0.14

obe

0.14

upo

0.14

lÃ©m

0.13

 Mist

0.13

opoulos

0.13

Activations Density 0.050%

elements related to publications and their editorial staff

No Comments

No Known Activations