INDEX

Explanations

specific references to the word "Reading" used as a proper noun

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "Reading."

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

xon

-0.93

ortality

-0.89

Ø§ÙĦ

-0.83

Thumbnail

-0.71

ounty

-0.66

rolet

-0.66

ascal

-0.66

ella

-0.66

avior

-0.65

pload

-0.64

POSITIVE LOGITS

 comprehension

1.30

 aloud

1.03

 glasses

0.88

just

0.86

 Reading

0.86

ied

0.82

Reading

0.76

reading

0.76

 readings

0.75

 reading

0.75

Activations Density 0.029%

specific references to the word "Reading" used as a proper noun

occurrences of the word "Reading."

No Comments

No Known Activations