INDEX

Explanations

phrases indicating a midpoint or space between two entities or concepts

oai_token-act-pair · gpt-3.5-turbo

instances of the word "between" and related concepts involving juxtaposition or division

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

vous

-0.76

dn

-0.76

apego

-0.75

cause

-0.75

tu

-0.73

DN

-0.71

gins

-0.69

rouse

-0.68

enegger

-0.68

overed

-0.67

POSITIVE LOGITS

heid

0.75

 form

0.66

 spheres

0.63

 vicinity

0.62

 trance

0.61

 bouts

0.60

 Territories

0.58

 shots

0.58

 drawer

0.58

 preparations

0.57

Activations Density 0.025%

phrases indicating a midpoint or space between two entities or concepts

instances of the word "between" and related concepts involving juxtaposition or division

No Comments

No Known Activations