INDEX

Explanations

bibliographic details of academic references

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

 Berm

-0.16

ersen

-0.15

jerne

-0.15

panse

-0.15

emachine

-0.14

stile

-0.14

 Shares

-0.14

rine

-0.13

EDITOR

-0.13

 Copper

-0.13

POSITIVE LOGITS

acle

0.16

umi

0.14

 prostitutas

0.14

irit

0.14

LayoutManager

0.14

 kvin

0.14

achten

0.13

 Ally

0.13

Î±Îº

0.13

indexOf

0.13

Activations Density 0.004%

bibliographic details of academic references

No Comments

No Known Activations