INDEX

Explanations

references to the past or nostalgia

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

woke

-0.17

aten

-0.17

Back

-0.15

Ø¡

-0.15

BACK

-0.15

 Back

-0.15

ought

-0.15

crate

-0.14

elsif

-0.14

 Later

-0.14

POSITIVE LOGITS

antha

0.18

 allo

0.17

ronym

0.17

InstanceState

0.17

 east

0.15

enus

0.14

AdapterFactory

0.14

/dd

0.14

loys

0.14

 eskort

0.14

Activations Density 0.009%

references to the past or nostalgia

No Comments

No Known Activations