INDEX

Explanations

negations and expressions of "not" 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 @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

-0.16

roll

-0.16

arak

-0.15

ceph

-0.15

ven

-0.15

illon

-0.15

Ire

-0.15

vers

-0.15

à¤¸à¤²

-0.14

å°Ķ

-0.14

POSITIVE LOGITS

osate

0.16

éĻ¶

0.16

IDA

0.16

ektor

0.15

ANGER

0.15

RICT

0.15

ida

0.15

ija

0.14

iesta

0.14

angkan

0.14

Activations Density 0.207%

negations and expressions of "not" in various contexts

No Comments

No Known Activations