INDEX

Explanations

mentions of the word 'comm' in the text

oai_token-act-pair · gpt-3.5-turbo

terms related to communication and its 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

 Elves

-0.66

 vulgar

-0.65

 Welsh

-0.65

geon

-0.63

 Tribe

-0.62

 Belg

-0.61

 Georgian

-0.61

 Corinth

-0.61

draw

-0.61

chest

-0.60

POSITIVE LOGITS

encing

1.17

otion

1.16

ittee

1.15

ences

1.14

ittees

1.12

mercial

1.12

essage

1.12

ittal

1.10

itte

1.09

onsense

1.08

Activations Density 0.024%

mentions of the word 'comm' in the text

terms related to communication and its processes

No Comments

No Known Activations