INDEX

Explanations

information related to graphs and charts, especially with descriptions of data and statistics

oai_token-act-pair · gpt-3.5-turbo

data representations and statistics related to rankings or charts

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

 Bapt

-0.77

 exorc

-0.70

assion

-0.68

anism

-0.66

 breathed

-0.66

 tant

-0.64

heimer

-0.64

edi

-0.64

 evangel

-0.63

 breat

-0.63

POSITIVE LOGITS

 dataset

1.02

 graphs

0.98

 spreadsheet

0.97

verages

0.94

 computed

0.94

 statistical

0.93

 probabilities

0.88

 snapshot

0.84

 graph

0.83

 weighted

0.82

Activations Density 0.705%

information related to graphs and charts, especially with descriptions of data and statistics

data representations and statistics related to rankings or charts

No Comments

No Known Activations