INDEX

Explanations

actions related to data manipulation and software operations

oai_token-act-pair · gpt-3.5-turbo

functional verbs related to software and technical processes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

reau

-0.73

backer

-0.72

quart

-0.71

study

-0.70

å§«

-0.68

enegger

-0.67

Tech

-0.66

 Study

-0.64

ingen

-0.64

\.

-0.62

POSITIVE LOGITS

 itself

1.04

 dynamically

0.82

its

0.76

ITS

0.73

 electrons

0.66

ibly

0.65

olve

0.65

lessly

0.64

 optimizations

0.63

ynchron

0.62

Activations Density 0.659%

actions related to data manipulation and software operations

functional verbs related to software and technical processes

No Comments

No Known Activations