INDEX

Explanations

phrases indicating spatial orientation or directionality

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

aghan

-0.19

Fat

-0.17

é«

-0.17

anio

-0.16

cad

-0.15

iangle

-0.15

tility

-0.15

Fat

-0.15

nila

-0.14

ailles

-0.14

POSITIVE LOGITS

cott

0.17

 Suff

0.16

uffer

0.15

 passage

0.14

ipes

0.14

tran

0.14

Æ°á»Ŀng

0.14

mut

0.14

 experiment

0.13

Paris

0.13

Activations Density 0.021%

phrases indicating spatial orientation or directionality

No Comments

No Known Activations