INDEX

Explanations

numeric measurements and comparisons related to documents or text length

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

 Lowe

-0.15

 Tarif

-0.15

 lateral

-0.14

elow

-0.13

LOW

-0.13

 sparing

-0.13

 timetable

-0.13

Lad

-0.13

ahat

-0.13

Lug

-0.13

POSITIVE LOGITS

 length

1.05

 Length

0.94

length

0.92

 lengths

0.92

-length

0.90

 longer

0.90

 lenght

0.85

Length

0.85

éķ¿åº¦

0.83

 longest

0.81

Activations Density 0.273%

numeric measurements and comparisons related to documents or text length

No Comments

No Known Activations