INDEX

Explanations

terms related to the immune system, such as immune cells, autoimmune, immune response, and antigen

oai_token-act-pair · gpt-3.5-turbo

terms related to the immune system

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

\\\\\\\\

-0.84

-0.81

ousand

-0.72

ilts

-0.72

 Paper

-0.71

 Rasmussen

-0.71

google

-0.70

ABE

-0.70

hare

-0.70

*/(

-0.69

POSITIVE LOGITS

 immune

0.95

mediated

0.82

 deficiency

0.81

 glob

0.79

 immunity

0.77

 tract

0.76

 reaction

0.75

 response

0.74

 detection

0.73

 defences

0.72

Activations Density 0.015%

terms related to the immune system, such as immune cells, autoimmune, immune response, and antigen

terms related to the immune system

No Comments

No Known Activations