INDEX

Explanations

terms related to biosciences and academic fields through detecting specific suffixes

oai_token-act-pair · gpt-3.5-turbo

references to "bios" or related academic terms

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

à©

-0.78

à¨

-0.77

ı

-0.76

thing

-0.69

 pains

-0.68

lished

-0.67

rying

-0.66

fare

-0.66

aneous

-0.62

dry

-0.62

POSITIVE LOGITS

aurus

1.19

pell

1.05

clerosis

1.01

omething

0.97

ophical

0.96

ystem

0.96

perm

0.93

erver

0.90

metics

0.88

chwitz

0.87

Activations Density 0.016%

terms related to biosciences and academic fields through detecting specific suffixes

references to "bios" or related academic terms

No Comments

No Known Activations