INDEX

Explanations

programming constructs related to namespaces and classes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 7-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L7R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.7.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

elerik

-0.15

agar

-0.15

.cgi

-0.15

sgi

-0.15

_PROC

-0.15

odo

-0.15

âĢ¦↵↵↵

-0.15

/Foundation

-0.14

/MIT

-0.14

SCRI

-0.14

POSITIVE LOGITS

0.15

ottes

0.15

th

0.15

 Ritual

0.15

 survived

0.14

_singleton

0.14

fre

0.14

ãĥ¼ãĥľ

0.14

Fre

0.14

Vig

0.14

Activations Density 0.001%

programming constructs related to namespaces and classes

No Comments

No Known Activations