INDEX

Explanations

terms associated with recognition or praise

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 5-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L5R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.5.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

è¨

-0.19

yll

-0.15

Ø§ÙģÙĬØ©

-0.14

ToBounds

-0.14

vise

-0.14

æ®¿

-0.13

íĪ¬

-0.13

 ÑįÑĤ

-0.13

etat

-0.13

POSITIVE LOGITS

ugar

0.17

ry

0.15

ertas

0.15

past

0.14

ascar

0.14

bane

0.14

Ø±Ùħ

0.14

 past

0.14

ermo

0.14

rome

0.13

Activations Density 0.040%

terms associated with recognition or praise

No Comments

No Known Activations