INDEX

Explanations

references to health-related conditions and factors

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

à¥įà¤ªà¤°

-0.16

í

-0.14

elan

-0.14

Sad

-0.13

 reels

-0.13

cursor

-0.13

çĵľ

-0.13

 conferred

-0.13

Jay

-0.13

POSITIVE LOGITS

 build

0.39

 buildup

0.35

 Build

0.35

build

0.33

 block

0.32

 jams

0.31

 builds

0.31

jam

0.30

Build

0.30

 stuck

0.29

Activations Density 0.151%

references to health-related conditions and factors

No Comments

No Known Activations