INDEX

Explanations

numeric data 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 @ 31-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L31R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.31.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

-0.15

ocale

-0.15

reich

-0.15

EIF

-0.14

ÏĮÎ´

-0.14

boro

-0.14

Raw

-0.14

 Ying

-0.14

ozo

-0.13

classed

-0.13

POSITIVE LOGITS

obile

0.13

/animate

0.13

Ã¼m

0.13

uliar

0.12

oute

0.12

 cres

0.12

Validity

0.11

voke

0.11

oney

0.11

_PROTOCOL

0.11

Activations Density 0.015%

numeric data or identifiers

No Comments

No Known Activations