INDEX

Explanations

financial data and information

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "total" in quantitative contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

©¶æ

-0.72

pher

-0.70

rir

-0.69

ĪĴ

-0.69

heid

-0.68

maid

-0.68

rium

-0.67

 Briggs

-0.66

bley

-0.66

otle

-0.62

POSITIVE LOGITS

itarian

1.08

 amount

0.86

itar

0.85

 number

0.78

 eclipse

0.77

 Recall

0.74

 elapsed

0.74

 Eclipse

0.73

 likelihood

0.72

 length

0.72

Activations Density 0.027%

financial data and information

occurrences of the word "total" in quantitative contexts

No Comments

No Known Activations