INDEX

Explanations

phrases and terms indicating limitations or restricted availability

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

roe

-0.16

-0.15

pok

-0.14

/she

-0.14

/her

-0.14

aviour

-0.14

oved

-0.14

ëľ

-0.14

neau

-0.13

ansi

-0.13

POSITIVE LOGITS

xa

0.17

åĽº

0.16

spb

0.16

ely

0.15

nun

0.15

ities

0.15

odore

0.14

ispens

0.14

ty

0.14

caps

0.14

Activations Density 0.043%

phrases and terms indicating limitations or restricted availability

No Comments

No Known Activations