INDEX

Explanations

prepositions or conjunctions used to introduce explanations or references

oai_token-act-pair · gpt-3.5-turbo

phrases indicating contexts of public discussion or discourse

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

hov

-0.87

romeda

-0.80

etheless

-0.75

BSD

-0.71

frog

-0.71

ochet

-0.71

issance

-0.70

dor

-0.69

oshi

-0.67

rolet

-0.66

POSITIVE LOGITS

 interviews

1.33

 speeches

1.30

 forums

1.29

 conversations

1.17

 polite

1.12

 debates

1.11

 contexts

1.11

 conversation

1.09

 public

1.08

 presentations

1.06

Activations Density 0.348%

prepositions or conjunctions used to introduce explanations or references

phrases indicating contexts of public discussion or discourse

No Comments

No Known Activations