INDEX

Explanations

references to the scientific journal "Nature"

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "Nature" and its variations

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

acters

-0.88

rons

-0.79

oning

-0.76

stan

-0.73

rano

-0.71

rican

-0.70

zai

-0.69

rations

-0.68

raph

-0.66

aged

-0.66

POSITIVE LOGITS

 Conserv

0.87

nature

0.80

 Nature

0.79

uces

0.75

Tree

0.74

 Neuroscience

0.70

 Plants

0.69

 preserves

0.68

+++

0.68

Nature

0.68

Activations Density 0.018%

references to the scientific journal "Nature"

occurrences of the word "Nature" and its variations

No Comments

No Known Activations