INDEX

Explanations

comparisons or contrasts being made between different entities or concepts

oai_token-act-pair · gpt-3.5-turbo

comparisons or contrasts between different subjects

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

 parliament

-0.75

WOOD

-0.70

ario

-0.66

RM

-0.65

ATA

-0.63

 Parliament

-0.62

ktop

-0.60

ornia

-0.60

igning

-0.59

redo

-0.58

POSITIVE LOGITS

MpServer

0.95

 favorably

0.86

ãĤ§

0.78

etheus

0.77

 dots

0.76

Ö¼

0.71

imilar

0.70

 CLSID

0.69

rimp

0.68

apse

0.67

Activations Density 0.184%

comparisons or contrasts being made between different entities or concepts

comparisons or contrasts between different subjects

No Comments

No Known Activations