INDEX

Explanations

terms related to electrical currents

oai_token-act-pair · gpt-3.5-turbo

terms related to electrical current

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

romeda

-0.91

alty

-0.82

Bei

-0.80

amy

-0.75

ËĪ

-0.72

 Aires

-0.71

qqa

-0.71

alties

-0.70

omy

-0.68

jah

-0.67

POSITIVE LOGITS

 predicament

0.77

 incarnation

0.76

 whereabouts

0.70

URRENT

0.70

 generation

0.69

Â¶

0.69

 condition

0.67

 iteration

0.67

 Directions

0.66

 pace

0.65

Activations Density 0.018%

terms related to electrical currents

terms related to electrical current

No Comments

No Known Activations