INDEX

Explanations

phrases related to efforts or attempts

oai_token-act-pair · gpt-3.5-turbo

occurrences of the phrase "to" followed by various actions or commitments

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

listed

-0.72

need

-0.71

lot

-0.68

points

-0.68

pict

-0.66

nin

-0.63

nces

-0.62

checks

-0.61

furt

-0.61

holes

-0.61

POSITIVE LOGITS

 achieve

1.01

 maximize

1.01

 promote

1.00

 improve

0.99

 accommodate

0.98

 solve

0.98

 resolve

0.96

 strengthen

0.96

 minimize

0.95

 overcome

0.95

Activations Density 0.546%

phrases related to efforts or attempts

occurrences of the phrase "to" followed by various actions or commitments

No Comments

No Known Activations