INDEX

Explanations

verbs related to causing or leading to a particular event or state

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.10.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

Ħ¢

-0.75

WER

-0.74

 cutter

-0.69

 compens

-0.67

 away

-0.64

 conve

-0.62

 Mechdragon

-0.60

 anat

-0.60

 dist

-0.60

 Hans

-0.59

POSITIVE LOGITS

ited

1.94

itated

1.93

itation

1.91

iting

1.87

itous

1.85

itating

1.85

itate

1.84

itations

1.82

itates

1.80

itable

1.80

Activations Density 0.075%

verbs related to causing or leading to a particular event or state

No Comments

No Known Activations