INDEX

Explanations

mathematical or computational terms related to sequences

oai_token-act-pair · gpt-3.5-turbo

terms related to sequences and ordered arrangements

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

ryn

-0.92

aghd

-0.84

lying

-0.81

sie

-0.76

isle

-0.75

enthal

-0.74

sburgh

-0.73

young

-0.73

sle

-0.72

nai

-0.71

POSITIVE LOGITS

 Sequence

1.10

 sequence

0.94

 sequences

0.87

ĸļ

0.85

 unfold

0.81

uality

0.80

 repeats

0.76

 unfolds

0.70

alysis

0.65

 weaving

0.65

Activations Density 0.017%

mathematical or computational terms related to sequences

terms related to sequences and ordered arrangements

No Comments

No Known Activations