INDEX

Explanations

references to web technologies and standards

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L1R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.1.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

iaux

-0.15

ÙĤØ¹

-0.14

aters

-0.14

 Strand

-0.14

 Infer

-0.14

Rus

-0.14

 Shadow

-0.14

Commerce

-0.14

orge

-0.14

bard

-0.13

POSITIVE LOGITS

NZ

0.15

Ø³Ø§ÙĨÛĮ

0.14

UNUSED

0.14

akra

0.14

anus

0.14

blink

0.14

arden

0.14

blade

0.14

lands

0.14

eworld

0.14

Activations Density 0.021%

references to web technologies and standards

No Comments

No Known Activations