INDEX

Explanations

words related to the concept of 'applied'

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "lied."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

 stride

-0.66

 shuttle

-0.64

 shorts

-0.63

 vacation

-0.63

 trunk

-0.63

 Task

-0.62

 walk

-0.61

 Google

-0.61

 segments

-0.60

 celebr

-0.60

POSITIVE LOGITS

lied

4.87

lying

1.86

lies

1.76

lie

1.40

PLIED

1.16

 Applied

1.11

stood

1.08

fab

1.05

ledged

1.04

went

1.03

Activations Density 0.010%

words related to the concept of 'applied'

occurrences of the word "lied."

No Comments

No Known Activations