INDEX

Explanations

references to percentages or statistical data

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 15-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L15R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.15.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

ous

-0.22

onn

-0.17

anda

-0.16

hot

-0.16

ala

-0.15

iff

-0.15

ãģĤãĤĭ

-0.15

ãĤīãģĹãģĦ

-0.15

aments

-0.15

lay

-0.14

POSITIVE LOGITS

tember

0.17

enaire

0.16

tile

0.16

ä»½

0.15

nbsp

0.15

eter

0.14

ÄħÅ¼

0.14

engin

0.14

entimes

0.14

iles

0.14

Activations Density 0.024%

references to percentages or statistical data

No Comments

No Known Activations

references to percentages or statistical data

No Comments

No Known Activations