INDEX

Explanations

keywords related to geographical locations - particularly boundaries, themes, and specific regions

oai_token-act-pair · gpt-3.5-turbo

terms related to boundaries, classifications, and specific categories

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

ĸļ

-0.88

 Beir

-0.68

anwhile

-0.67

 Debor

-0.62

 Townsend

-0.59

enegger

-0.59

 Rober

-0.58

 Exhibition

-0.57

senal

-0.57

bsite

-0.56

POSITIVE LOGITS

ÂŃ

0.60

related

0.55

 flavors

0.55

isations

0.55

offic

0.54

icons

0.53

secut

0.53

 servic

0.51

imes

0.51

 categories

0.51

Activations Density 0.671%

keywords related to geographical locations - particularly boundaries, themes, and specific regions

terms related to boundaries, classifications, and specific categories

No Comments

No Known Activations