INDEX

Explanations

references to layout attributes in programming or design environments

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

 hybrids

-0.15

olith

-0.14

inge

-0.14

ades

-0.14

hta

-0.14

èĭ

-0.14

è£ķ

-0.13

es

-0.13

 witnesses

-0.13

pta

-0.13

POSITIVE LOGITS

stroy

0.16

ynet

0.15

bab

0.15

0.14

ldr

0.14

-chan

0.14

IQUE

0.14

ayo

0.14

 molec

0.13

BAB

0.13

Activations Density 0.112%

references to layout attributes in programming or design environments

No Comments

No Known Activations