INDEX

Explanations

pronouns indicating subjects in sentences

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

 door

-0.16

çĭĲ

-0.15

wed

-0.14

ROWN

-0.14

oug

-0.14

UCK

-0.14

ÏĦÎ·

-0.14

 Ã¶l

-0.14

aster

-0.14

POSITIVE LOGITS

ymology

0.16

 ÑĢÑĸÑĩ

0.15

BERS

0.14

endir

0.14

äº

0.14

ker

0.14

 Affairs

0.14

çĴĥ

0.14

ovice

0.14

à¸²à¸«à¸¥

0.14

Activations Density 0.038%

pronouns indicating subjects in sentences

No Comments

No Known Activations