INDEX

Explanations

various forms of structured data or hierarchical relationships in content

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

OPSIS

-0.16

ned

-0.15

 Remed

-0.15

alis

-0.15

ãĤ·ãĥ£ãĥ«

-0.14

PLICIT

-0.14

quiv

-0.14

.Std

-0.14

wend

-0.14

POSITIVE LOGITS

ingle

0.17

athers

0.17

elan

0.16

essim

0.15

á»³

0.15

bag

0.15

«

0.15

ÐµÐ½ÐºÐ°

0.15

ari

0.15

 calendars

0.14

Activations Density 0.030%

various forms of structured data or hierarchical relationships in content

No Comments

No Known Activations

various forms of structured data or hierarchical relationships in content

No Comments

No Known Activations