INDEX

Explanations

time-related indicators and numerical data

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 2-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L2R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.2.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

ute

-0.17

UTE

-0.16

imizer

-0.16

ouri

-0.15

icode

-0.15

 Funk

-0.14

ãĤ¯ãĥĪ

-0.14

rell

-0.14

 Tess

-0.14

ilton

-0.14

POSITIVE LOGITS

illard

0.15

ochen

0.15

qual

0.15

 acquaintance

0.15

oola

0.14

eum

0.14

 reins

0.14

ESA

0.14

 qualifiers

0.14

 autonom

0.13

Activations Density 0.037%

time-related indicators and numerical data

No Comments

No Known Activations