INDEX

Explanations

repeated mentions of the word "we" in reference to collective actions or sentiments

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

Ð¸

-0.14

ÏĥÏĦÏģÎ¿

-0.14

aid

-0.14

-bin

-0.14

omen

-0.13

dao

-0.13

ÙħÙĬÙĦ

-0.13

SizeMode

-0.13

å¥

-0.13

acet

-0.13

POSITIVE LOGITS

kus

0.15

ierz

0.14

_DC

0.14

nings

0.13

asso

0.13

Ð±ÐµÑĢ

0.13

actable

0.13

Ø§Ùĩ

0.13

variant

0.13

Activations Density 0.017%

repeated mentions of the word "we" in reference to collective actions or sentiments

No Comments

No Known Activations