INDEX

Explanations

phrases indicating a comparison or evaluation of situations or entities

oai_token-act-pair · gpt-3.5-turbo

phrases emphasizing multiple perspectives or aspects of a subject

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

-0.71

 Saud

-0.70

aband

-0.66

etts

-0.65

itton

-0.65

ansas

-0.63

arez

-0.62

agra

-0.62

iere

-0.60

rade

-0.59

POSITIVE LOGITS

 resembles

0.95

 resemble

0.81

places

0.76

 resembled

0.75

hops

0.75

terness

0.71

 horr

0.70

 resemb

0.68

 surpr

0.67

 somew

0.66

Activations Density 0.053%

phrases indicating a comparison or evaluation of situations or entities

phrases emphasizing multiple perspectives or aspects of a subject

No Comments

No Known Activations