INDEX

Explanations

statements expressing obligation or necessity

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

aille

-0.17

isas

-0.17

utex

-0.15

áº¯t

-0.15

anas

-0.15

peare

-0.15

embro

-0.15

tributes

-0.14

croft

-0.14

Æ°á»Ľ

-0.14

POSITIVE LOGITS

ahn

0.17

_fence

0.16

.must

0.16

 Weekly

0.15

REP

0.14

Transpose

0.14

 variation

0.14

ampo

0.14

avanaugh

0.14

Activations Density 0.063%

statements expressing obligation or necessity

No Comments

No Known Activations