INDEX

Explanations

references and information related to art, photography, and specific events and locations

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs3072-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

3,072

Data Type

torch.float32

Hook Point

blocks.8.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ensibly

-0.80

ãĥ¼ãĥĨãĤ£

-0.76

\":

-0.69

ãĥ´ãĤ¡

-0.67

 tremend

-0.67

itely

-0.65

ãĥĲ

-0.63

 shred

-0.63

ãĥ©ãĥ³

-0.62

athered

-0.61

POSITIVE LOGITS

<|endoftext|>

1.25

 =================================

1.11

 =================================================================

1.06

 ----------------------------------------------------------------

1.04

 --------------------------------

1.03

1.01

↵

1.01

↵↵

1.00

↵Âł

0.96

Activations Density 1.948%

references and information related to art, photography, and specific events and locations

No Comments

No Known Activations