INDEX

Explanations

conjunctions and phrases indicating connections or relationships between subjects

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 5-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L5R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.5.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

 axis

-0.14

ullah

-0.14

ON

-0.14

owi

-0.13

 Markus

-0.13

 gener

-0.13

.jpeg

-0.13

HC

-0.13

spot

-0.13

gh

-0.13

POSITIVE LOGITS

CJK

0.15

/Gate

0.14

Jub

0.14

alte

0.14

Ã¤ge

0.14

MBER

0.14

oplay

0.14

.untracked

0.14

ober

0.14

/scripts

0.14

Activations Density 0.045%

conjunctions and phrases indicating connections or relationships between subjects

No Comments

No Known Activations