INDEX

Explanations

words related to foreseeable events or situations

oai_token-act-pair · gpt-3.5-turbo

terms related to the concept of predictability and forecasting

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

cius

-0.94

rosse

-0.83

rams

-0.77

cart

-0.74

apa

-0.74

olding

-0.73

rooms

-0.72

inance

-0.72

raid

-0.69

por

-0.69

POSITIVE LOGITS

 foreseeable

1.28

($)

0.80

 unlaw

0.80

ĸļ

0.77

ãĤ¼ãĤ¦ãĤ¹

0.77

 minded

0.77

æĹ

0.75

 conclud

0.74

 ãĤµãĥ¼ãĥĨãĤ£ãĥ¯ãĥ³

0.74

ãĤ¦

0.71

Activations Density 0.004%

words related to foreseeable events or situations

terms related to the concept of predictability and forecasting

No Comments

No Known Activations