INDEX

Explanations

facts or figures that exceed a certain threshold

oai_token-act-pair · gpt-3.5-turbo

terms related to surpassing or exceeding thresholds or limits

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

atto

-0.67

uzzle

-0.65

SO

-0.63

rug

-0.63

rain

-0.62

iola

-0.61

away

-0.61

abet

-0.60

 rotated

-0.60

rounder

-0.59

POSITIVE LOGITS

 expectations

1.00

atos

0.86

0.80

ingly

0.76

whelming

0.73

=>

0.72

İĭ

0.70

 orgasm

0.70

hap

0.68

ealous

0.66

Activations Density 0.076%

facts or figures that exceed a certain threshold

terms related to surpassing or exceeding thresholds or limits

No Comments

No Known Activations