INDEX

Explanations

instances of the word "full" and related terms, particularly in contexts involving resources or comprehensive content

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 6-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L6R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.6.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

orias

-0.16

 Wade

-0.15

 Provid

-0.14

arty

-0.14

aders

-0.14

edo

-0.14

lock

-0.14

 ******************************************************************************↵

-0.14

 Phelps

-0.14

 Forward

-0.13

POSITIVE LOGITS

ernes

0.19

ERT

0.16

ç«ł

0.15

erto

0.14

rete

0.14

.TableName

0.14

istine

0.14

 Nack

0.14

NTN

0.14

ÙĩØ¯Ø§Ùģ

0.13

Activations Density 0.007%

instances of the word "full" and related terms, particularly in contexts involving resources or comprehensive content

No Comments

No Known Activations