INDEX

Explanations

adjectives describing the emotional quality or attitude of a statement

oai_token-act-pair · gpt-3.5-turbo

references to tone in discussions or narratives

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

minster

-0.77

icipated

-0.76

lisher

-0.74

riks

-0.73

ernels

-0.72

iott

-0.71

essions

-0.71

alus

-0.70

arus

-0.69

actus

-0.68

POSITIVE LOGITS

 deaf

1.17

 tone

1.01

 Tone

0.91

 assurance

0.82

 tones

0.77

 tune

0.75

 demeanor

0.71

 dial

0.70

 shif

0.68

 amplification

0.68

Activations Density 0.036%

adjectives describing the emotional quality or attitude of a statement

references to tone in discussions or narratives

No Comments

No Known Activations