INDEX

Explanations

phrases or references to significant events or details

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L1R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.1.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

aram

-0.19

anic

-0.15

 voks

-0.15

.uf

-0.15

RPC

-0.15

audi

-0.15

Press

-0.14

vet

-0.14

linger

-0.14

 Caval

-0.14

POSITIVE LOGITS

ooke

0.17

 dopad

0.16

issan

0.16

Sink

0.14

uisine

0.14

ryptography

0.14

.Void

0.13

 originally

0.13

inqu

0.13

oren

0.13

Activations Density 0.299%

phrases or references to significant events or details

No Comments

No Known Activations