INDEX

Explanations

instances of the word "with" and its variations in context

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

ropol

-0.16

erland

-0.15

ibri

-0.14

apon

-0.14

bote

-0.14

_sink

-0.14

inize

-0.14

Įĵ

-0.14

egin

-0.14

uchar

-0.14

POSITIVE LOGITS

ita

0.15

obus

0.15

iko

0.14

kiego

0.14

ABEL

0.14

Hi

0.14

dued

0.14

ki

0.13

hma

0.13

Activations Density 0.012%

instances of the word "with" and its variations in context

No Comments

No Known Activations