INDEX

Explanations

specific identifiers, likely related to unique records or entries in a database

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 7-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L7R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.7.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

cente

-0.15

iens

-0.15

hung

-0.14

itto

-0.14

colo

-0.14

emoc

-0.14

 Severity

-0.14

ë¡

-0.14

uppen

-0.14

ebek

-0.14

POSITIVE LOGITS

unb

0.16

 trat

0.15

ANDOM

0.14

 myself

0.14

.LA

0.14

’

0.14

tay

0.13

 Heller

0.13

 Ø§ÙĦÙĬ

0.13

Activations Density 0.013%

specific identifiers, likely related to unique records or entries in a database

No Comments

No Known Activations