INDEX

Explanations

references to visual media and entertainment forms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

áº¿p

-0.16

itesse

-0.16

RIA

-0.15

ãĥ¼ãĥĢ

-0.14

Ø±ÙĪÙħ

-0.14

eca

-0.14

=back

-0.14

ffen

-0.14

stantiate

-0.14

(OS

-0.14

POSITIVE LOGITS

llib

0.18

jer

0.15

angen

0.14

Ãłng

0.14

 Ethi

0.14

ovenant

0.14

éº»

0.14

wers

0.14

Beg

0.14

 domic

0.14

Activations Density 0.250%

references to visual media and entertainment forms

No Comments

No Known Activations