INDEX

Explanations

phrases that indicate personal agreement or alignment with opinions and assessments

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs49152-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

49,152

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

ishable

-0.72

ificial

-0.70

 resil

-0.69

 Fior

-0.66

 unsuspecting

-0.64

andise

-0.62

hidden

-0.62

 staffed

-0.62

ornia

-0.61

Joined

-0.60

POSITIVE LOGITS

 tenets

1.00

 premise

0.94

 conclusions

0.93

 proposition

0.88

 wording

0.87

 sentiments

0.87

 principle

0.87

 characterization

0.86

 thesis

0.84

 Terms

0.81

Activations Density 0.094%

phrases that indicate personal agreement or alignment with opinions and assessments

No Comments

No Known Activations