INDEX

Explanations

conversations or dialogues involving individuals expressing assertive or confrontational behaviors

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_fs24576-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

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

 hiber

-0.81

 indo

-0.70

 disg

-0.68

 rejuven

-0.66

oun

-0.65

 skelet

-0.65

 helicop

-0.64

 mathemat

-0.64

hovah

-0.63

 crochet

-0.63

POSITIVE LOGITS

%:

1.07

Response

1.05

Reply

1.01

':

1.00

Others

0.98

*:

0.94

Reason

0.93

]:

0.93

Result

0.92

Solution

0.92

Activations Density 0.167%

conversations or dialogues involving individuals expressing assertive or confrontational behaviors

No Comments

No Known Activations