INDEX

Explanations

lines or sections with placeholder or formatting symbols used in forms or documents

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

anas

-0.17

iras

-0.15

 Clair

-0.14

atar

-0.14

ableObject

-0.14

zbollah

-0.14

à¸´à¹Ĥ

-0.14

upy

-0.13

 escal

-0.13

ement

-0.13

POSITIVE LOGITS

ship

0.17

-translate

0.15

ighth

0.15

iese

0.15

ìĸµ

0.15

.toObject

0.14

.GetBytes

0.14

 pity

0.14

constitutional

0.13

adget

0.13

Activations Density 0.023%

lines or sections with placeholder or formatting symbols used in forms or documents

No Comments

No Known Activations