INDEX

Explanations

location-related details in a text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

.sap

-0.15

kyt

-0.15

fÃ¼h

-0.15

Colorado

-0.14

ç¬¦

-0.14

onaut

-0.13

aud

-0.13

Gon

-0.13

(++

-0.13

vm

-0.13

POSITIVE LOGITS

 Liverpool

0.51

 Liver

0.47

Liverpool

0.46

 liver

0.44

Liver

0.42

Wir

0.35

Mer

0.34

 Everton

0.33

hep

0.31

Wir

0.31

Activations Density 0.022%

location-related details in a text

No Comments

No Known Activations