INDEX

Explanations

the word "as" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

iyim

-0.14

orts

-0.14

igrations

-0.14

 allerdings

-0.14

ondon

-0.14

ippers

-0.14

ably

-0.14

sys

-0.13

omens

-0.13

aÃ§

-0.13

POSITIVE LOGITS

 there

0.28

 opposed

0.23

there

0.23

 none

0.21

 regards

0.19

 There

0.19

it

0.18

 oppose

0.17

There

0.17

 unlike

0.17

Activations Density 0.100%

the word "as" in various contexts

No Comments

No Known Activations