INDEX

Explanations

mentions of the word "first" along with related sequence or order terms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 7-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L7R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.7.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

-0.15

inan

-0.14

Uh

-0.14

ombo

-0.14

imp

-0.14

SAN

-0.14

San

-0.14

POSITIVE LOGITS

 Olsen

0.16

abra

0.15

/latest

0.15

-last

0.15

ãĤ¤ãĤº

0.15

abs

0.15

anni

0.15

ÎµÎ¹Î¿

0.15

ABS

0.15

antage

0.15

Activations Density 0.074%

mentions of the word "first" along with related sequence or order terms

No Comments

No Known Activations