INDEX

Explanations

special characters or symbols, likely focused on specific formatting or encoded information

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

.private

-0.16

 Took

-0.15

neau

-0.15

orex

-0.15

alytics

-0.15

 LENG

-0.15

ÄĽÅ¾

-0.15

$MESS

-0.14

ayd

-0.14

bay

-0.14

POSITIVE LOGITS

 seeming

0.20

 appear

0.20

 seem

0.18

 appears

0.18

 seemed

0.17

 appeared

0.17

Appear

0.17

 appearing

0.17

.UnitTesting

0.17

 seems

0.16

Activations Density 0.006%

special characters or symbols, likely focused on specific formatting or encoded information

No Comments

No Known Activations