INDEX

Explanations

keywords related to evaluation and recommendations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

Bau

-0.18

Ã¢u

-0.16

achts

-0.16

Ð°ÑĤÐºÐ¾Ð²

-0.16

makta

-0.15

Î±Î½Î±

-0.15

ÃŃg

-0.15

maktan

-0.15

 Benson

-0.14

POSITIVE LOGITS

able

0.95

ability

0.71

ables

0.70

ABLE

0.69

ably

0.66

ible

0.61

abil

0.57

ble

0.56

abled

0.55

abl

0.54

Activations Density 0.157%

keywords related to evaluation and recommendations

No Comments

No Known Activations