INDEX

Explanations

URLs and references to issues or tickets

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

imbledon

-0.16

eka

-0.16

foy

-0.13

 Laden

-0.13

ware

-0.13

olumn

-0.13

Î¹Î¸

-0.13

INavigation

-0.13

tet

-0.13

POSITIVE LOGITS

ocale

0.15

jango

0.15

obuf

0.14

arrison

0.14

asl

0.14

arcy

0.14

achi

0.14

avin

0.14

ben

0.14

oran

0.14

Activations Density 0.047%

URLs and references to issues or tickets

No Comments

No Known Activations