INDEX

Explanations

references to audiovisual content and associated metadata

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

à¤²à¤Ĺ

-0.16

uc

-0.15

vinces

-0.14

à¥įà¤¤à¤µ

-0.14

 convers

-0.14

etheus

-0.14

pid

-0.14

lore

-0.13

itrust

-0.13

_ment

-0.13

POSITIVE LOGITS

 Kami

0.14

ouser

0.13

ehen

0.13

ãĥĥãĤ«ãĥ¼

0.13

isi

0.13

å¹²

0.13

ousse

0.13

ices

0.13

 ÐĿÐ°Ð´

0.13

Activations Density 0.018%

references to audiovisual content and associated metadata

No Comments

No Known Activations