INDEX

Explanations

references to time, specifically focusing on the word "past."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 2-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L2R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.2.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

zew

-0.16

ohn

-0.15

_SPI

-0.15

ubo

-0.14

ane

-0.14

ubat

-0.14

Ãºb

-0.13

buz

-0.13

ickey

-0.13

ÙĳÙĦ

-0.13

POSITIVE LOGITS

lama

0.15

lava

0.15

ith

0.15

.fm

0.15

rong

0.14

.UnitTesting

0.14

arsers

0.14

_putchar

0.14

DCF

0.14

ictures

0.14

Activations Density 0.012%

references to time, specifically focusing on the word "past."

No Comments

No Known Activations