INDEX

Explanations

visual representations of data

oai_token-act-pair · gpt-3.5-turbo

information related to visual representations or graphical data in scientific contexts

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

 wiser

-0.75

angan

-0.72

ensable

-0.71

 unwitting

-0.69

antic

-0.69

addon

-0.68

bred

-0.68

 rightful

-0.67

 incest

-0.67

obbies

-0.67

POSITIVE LOGITS

 graph

0.93

 depicts

0.91

 Figure

0.90

 graphs

0.90

below

0.89

Figure

0.86

 square

0.83

 below

0.82

 circles

0.82

 illustration

0.81

Activations Density 0.306%

visual representations of data

information related to visual representations or graphical data in scientific contexts

No Comments

No Known Activations