INDEX

Explanations

phrases indicating a high likelihood or probability of future events

oai_token-act-pair · gpt-3.5-turbo

phrases indicating probable outcomes or predictions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

rief

-0.78

ente

-0.77

inth

-0.76

olded

-0.72

aan

-0.72

oos

-0.71

gado

-0.70

 Tags

-0.70

gian

-0.70

entric

-0.69

POSITIVE LOGITS

 releg

0.79

cffff

0.78

 ingred

0.70

 elim

0.69

 likely

0.69

 unanimous

0.69

 infer

0.68

 elector

0.68

 confir

0.67

ãĥ´

0.66

Activations Density 0.023%

phrases indicating a high likelihood or probability of future events

phrases indicating probable outcomes or predictions

No Comments

No Known Activations