INDEX

Explanations

connections and associations within various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

_relations

-0.17

yle

-0.16

Relations

-0.16

ptions

-0.16

 Relations

-0.16

GLOBALS

-0.15

 relations

-0.15

ses

-0.15

_relationship

-0.15

rial

-0.15

POSITIVE LOGITS

ness

0.27

LY

0.17

ly

0.17

èģĶ

0.17

-party

0.16

odore

0.16

ational

0.16

 cháº·t

0.16

ëĲľ

0.15

 expenses

0.15

Activations Density 0.041%

connections and associations within various contexts

No Comments

No Known Activations