INDEX

Explanations

terms related to neurotransmitters and behavioral changes

oai_token-act-pair · gpt-3.5-turbo

terms related to behavioral and psychological studies

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

shit

-0.83

Advertisement

-0.82

 piss

-0.81

Kids

-0.80

pee

-0.79

bedroom

-0.78

 Parkway

-0.77

 booze

-0.76

TextColor

-0.75

ulhu

-0.74

POSITIVE LOGITS

 âĪ¼

1.32

 causal

1.24

 endogenous

1.22

 heterogeneity

1.18

 caus

1.18

 phenotype

1.16

 hypothesized

1.15

 methodological

1.14

 hypotheses

1.14

 eluc

1.09

Activations Density 0.959%

terms related to neurotransmitters and behavioral changes

terms related to behavioral and psychological studies

No Comments

No Known Activations