INDEX

Explanations

numerals describing measurement units

oai_token-act-pair · gpt-3.5-turbo

references to measurements and specifications, particularly involving quotes and sizes

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

frey

-0.74

ibal

-0.71

odor

-0.71

atically

-0.69

akespe

-0.68

abeth

-0.68

 Selected

-0.67

 recapt

-0.66

ophen

-0.65

breeding

-0.65

POSITIVE LOGITS

 diameter

0.91

 tall

0.79

itudinal

0.75

 thickness

0.73

 increments

0.72

PVC

0.71

 thick

0.71

 inch

0.69

 angle

0.68

 joints

0.68

Activations Density 0.036%

numerals describing measurement units

references to measurements and specifications, particularly involving quotes and sizes

No Comments

No Known Activations