INDEX

Explanations

terms related to psychiatric conditions

oai_token-act-pair · gpt-3.5-turbo

references to psychiatric topics and conditions

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

Cu

-0.70

layer

-0.70

Mo

-0.68

 beads

-0.68

 arrows

-0.67

fficiency

-0.67

Kick

-0.66

 pots

-0.65

 friction

-0.65

Cu

-0.65

POSITIVE LOGITS

 psychiatric

3.14

 psychiatry

2.75

 Psychiatric

2.58

 psychiatrists

2.41

 Psychiatry

2.19

 psychiat

2.14

 psychiatrist

2.09

 Psychiat

1.89

DSM

1.81

 antidepressants

1.79

Activations Density 0.027%

terms related to psychiatric conditions

references to psychiatric topics and conditions

No Comments

No Known Activations