INDEX

Explanations

repetitive or nonsensical phrases that seem out of context or unusual in the text

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

orah

-0.82

ossibility

-0.79

ittee

-0.73

owler

-0.70

fecture

-0.69

ocese

-0.68

leneck

-0.68

utenant

-0.68

former

-0.67

aucus

-0.67

POSITIVE LOGITS

 amounts

1.38

 quantities

1.28

 versions

1.23

 doses

1.15

 solutions

1.13

 scenarios

1.13

 messages

1.09

 situations

1.08

 ones

1.07

 things

1.07

Activations Density 0.307%

repetitive or nonsensical phrases that seem out of context or unusual in the text

No Comments

No Known Activations