INDEX

Explanations

numeric quantities or totals

oai_token-act-pair · gpt-3.5-turbo

the occurrence of the word "total" and its associated numerical values

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ĪĴ

-0.72

heid

-0.71

©¶æ

-0.69

rir

-0.67

sole

-0.66

roe

-0.66

 Racer

-0.65

bley

-0.64

ste

-0.64

maid

-0.63

POSITIVE LOGITS

itarian

1.04

 amount

0.92

itar

0.84

 number

0.82

 Estimated

0.77

 elapsed

0.76

 length

0.75

 allowable

0.74

 eclipse

0.74

 compensation

0.73

Activations Density 0.040%

numeric quantities or totals

the occurrence of the word "total" and its associated numerical values

No Comments

No Known Activations