INDEX

Explanations

instances of numerical data or specific schedules

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

 January

-0.18

 February

-0.18

 äºİ

-0.17

January

-0.16

 March

-0.16

February

-0.16

March

-0.16

spring

-0.16

 April

-0.15

 July

-0.15

POSITIVE LOGITS

æĺŁæľŁ

0.21

0.19

0.18

vetica

0.18

 exactly

0.15

erdale

0.14

jl

0.14

istrovstvÃŃ

0.14

 exact

0.14

oust

0.14

Activations Density 0.058%

instances of numerical data or specific schedules

No Comments

No Known Activations