INDEX

Explanations

information related to changes and conditions in various contexts, particularly focusing on metrics or measurements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

vens

-0.17

Ð°ÐºÐ°

-0.16

 italia

-0.15

 acronym

-0.15

ala

-0.14

ÏĨÎ¿

-0.14

Ø§ÙĦØ§

-0.14

chron

-0.14

rief

-0.14

oca

-0.14

POSITIVE LOGITS

 Bard

0.16

orris

0.16

abin

0.15

ILLISECONDS

0.15

apia

0.15

ISR

0.14

ILLISE

0.14

odash

0.14

ude

0.14

uden

0.14

Activations Density 1.230%

information related to changes and conditions in various contexts, particularly focusing on metrics or measurements

No Comments

No Known Activations