INDEX

Explanations

references to the United States and its context in various subjects

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

APER

-0.17

elf

-0.17

ing

-0.15

erspective

-0.14

ety

-0.14

ften

-0.14

rung

-0.14

ie

-0.14

hind

-0.14

 sensitive

-0.14

POSITIVE LOGITS

/world

0.17

-wide

0.16

wide

0.16

merican

0.15

PTS

0.15

Enlarge

0.15

progressbar

0.14

Wide

0.14

<source

0.14

 dime

0.14

Activations Density 0.019%

references to the United States and its context in various subjects

No Comments

No Known Activations