INDEX

Explanations

elements and attributes related to markup or coding structures

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 16-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L16R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.16.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

 Gould

-0.16

Til

-0.15

iske

-0.15

arcer

-0.15

otland

-0.15

olla

-0.14

lesia

-0.14

 Grove

-0.13

elden

-0.13

 ç¢

-0.13

POSITIVE LOGITS

ewire

0.15

ÅĻel

0.14

 incom

0.14

 poil

0.13

irket

0.13

 keywords

0.13

 elev

0.13

debit

0.13

Instr

0.13

Activations Density 0.027%

elements and attributes related to markup or coding structures

No Comments

No Known Activations