INDEX

Explanations

repeated mentions of the word "sam" or its variations, indicating a focus on character names or identifiers

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

ÂĽ

-0.18

anel

-0.17

iba

-0.15

ansa

-0.14

isclosed

-0.14

ULSE

-0.14

ust

-0.13

 Confidence

-0.13

 Murdoch

-0.13

usta

-0.13

POSITIVE LOGITS

0.17

ex

0.16

bou

0.15

inos

0.14

à§įà¦

0.14

fv

0.14

Ã³rio

0.14

dana

0.14

urger

0.14

Activations Density 0.023%

repeated mentions of the word "sam" or its variations, indicating a focus on character names or identifiers

No Comments

No Known Activations