INDEX

Explanations

terms related to collaboration 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

ColumnsMode

-0.16

yen

-0.15

listen

-0.15

ãģĵãģĿ

-0.15

longleftrightarrow

-0.15

rist

-0.14

emark

-0.14

dem

-0.14

edException

-0.14

emen

-0.14

POSITIVE LOGITS

abor

0.29

ateral

0.23

ared

0.21

isions

0.21

aps

0.21

[Byte

0.21

 coll

0.21

agen

0.20

apsed

0.20

apsible

0.19

Activations Density 0.010%

terms related to collaboration within various contexts

No Comments

No Known Activations