INDEX

Explanations

phrases containing the word "mid" followed by a number

oai_token-act-pair · gpt-3.5-turbo

references to the concept of "mid" as in midpoints or mid-decades

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

edIn

-0.74

Magikarp

-0.67

 Bearing

-0.66

 Redemption

-0.63

 Immunity

-0.63

itsch

-0.62

Reviewed

-0.62

 Benef

-0.61

çİĭ

-0.60

 Preferred

-0.60

POSITIVE LOGITS

-'

0.91

cellaneous

0.91

iors

0.88

season

0.88

ummer

0.87

wives

0.86

terms

0.86

west

0.84

western

0.83

century

0.81

Activations Density 0.013%

phrases containing the word "mid" followed by a number

references to the concept of "mid" as in midpoints or mid-decades

No Comments

No Known Activations