INDEX

Explanations

multitudinous mentions of the word "to"

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

nikov

-0.70

 Factors

-0.70

 Adviser

-0.68

(<

-0.65

[+

-0.65

 Advertisement

-0.65

ivil

-0.64

 inflamm

-0.64

 Explan

-0.63

Refer

-0.63

POSITIVE LOGITS

 fruition

1.13

 grips

0.81

 haunt

0.79

 Vegas

0.79

 shelves

0.76

 Clarks

0.73

 vinyl

0.73

 Louisville

0.72

 sweep

0.72

 shores

0.72

Activations Density 0.058%

multitudinous mentions of the word "to"

No Comments

No Known Activations