INDEX

Explanations

time-related information

oai_token-act-pair · gpt-3.5-turbo

numerical totals and measurements

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

inger

-0.66

 spat

-0.63

Myr

-0.61

 spiked

-0.61

 Mate

-0.60

eller

-0.59

 centuries

-0.58

reprene

-0.58

cients

-0.56

 Brist

-0.56

POSITIVE LOGITS

ngth

0.77

VALUE

0.77

urden

0.75

minus

0.74

 expenditures

0.74

 Expend

0.72

 expenditure

0.70

 liabilities

0.69

Required

0.68

ãĥ¯ãĥ³

0.67

Activations Density 0.209%

time-related information

numerical totals and measurements

No Comments

No Known Activations