INDEX

Explanations

segments or parts of a series

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

ermen

-0.17

ç·Ĵ

-0.16

 passive

-0.16

ophy

-0.15

 Wayback

-0.15

erb

-0.15

opes

-0.14

 Cant

-0.14

POSITIVE LOGITS

-series

0.17

Äł

0.16

 addCriterion

0.16

ousel

0.16

rego

0.15

.series

0.15

 KrÃ¡l

0.15

Locker

0.15

ichick

0.15

pper

0.15

Activations Density 0.050%

segments or parts of a series

No Comments

No Known Activations