INDEX

Explanations

phrases related to geographic locations and urban infrastructure

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

avity

-0.15

ieber

-0.15

orelease

-0.14

orado

-0.14

reb

-0.14

unker

-0.14

osh

-0.14

éĻµ

-0.14

 Bedrooms

-0.14

-src

-0.14

POSITIVE LOGITS

 walk

0.96

 Walk

0.90

walk

0.88

 walks

0.88

 walking

0.87

Walk

0.84

 walked

0.82

.walk

0.79

 walkers

0.78

 walker

0.77

Activations Density 0.353%

phrases related to geographic locations and urban infrastructure

No Comments

No Known Activations