INDEX

Explanations

phrases related to acquisition or obtaining items or information

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

askan

-0.20

ackers

-0.16

iÅŁ

-0.16

oÄį

-0.15

Ã¢u

-0.15

usc

-0.14

,readonly

-0.14

.tele

-0.14

 retention

-0.14

(utf

-0.14

POSITIVE LOGITS

 copy

0.17

 copies

0.16

 edin

0.16

iego

0.15

.async

0.15

 Merr

0.14

esis

0.14

/effects

0.14

 Copy

0.14

Activations Density 0.002%

phrases related to acquisition or obtaining items or information

No Comments

No Known Activations