INDEX

Explanations

words related to measurement devices, specifically 'ometer'

oai_token-act-pair · gpt-3.5-turbo

terms related to various types of measuring devices

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

lez

-0.77

giving

-0.71

ership

-0.69

ABE

-0.68

ggies

-0.68

Kut

-0.67

edu

-0.67

don

-0.65

clus

-0.64

ãģį

-0.63

POSITIVE LOGITS

ometer

0.85

 calibration

0.78

 measuring

0.77

 readings

0.75

ium

0.74

 calibrated

0.72

apter

0.70

 buoy

0.70

 tampering

0.69

 malfunction

0.69

Activations Density 0.029%

words related to measurement devices, specifically 'ometer'

terms related to various types of measuring devices

No Comments

No Known Activations