INDEX

Explanations

references to computational concepts and processes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 31-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L31R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.31.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

ahir

-0.14

 Santana

-0.14

berman

-0.13

.owl

-0.12

Truthy

-0.12

 Ø§ÙĦÙĨÙĪ

-0.12

£p

-0.12

TimeInterval

-0.12

aggable

-0.12

é«

-0.12

POSITIVE LOGITS

FOR

0.85

FOR

0.84

For

0.84

 Forrest

0.80

 Ð¤Ð¾ÑĢ

0.79

For

0.78

 Fore

0.77

-for

0.76

_For

0.76

_FOR

0.74

Activations Density 0.215%

references to computational concepts and processes

No Comments

No Known Activations