INDEX

Explanations

statements related to investigations

oai_token-act-pair · gpt-3.5-turbo

instances of investigations or inquiries being conducted

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

ãĤª

-0.72

hold

-0.69

eries

-0.64

âĹ¼

-0.64

erie

-0.62

 goodbye

-0.60

 farewell

-0.60

birth

-0.60

held

-0.59

emonic

-0.59

POSITIVE LOGITS

 allegations

0.84

 wrongdoing

0.80

 whether

0.76

 discl

0.73

 investigates

0.71

 alleged

0.71

 independently

0.70

 probing

0.70

agate

0.70

 UFOs

0.69

Activations Density 0.039%

statements related to investigations

instances of investigations or inquiries being conducted

No Comments

No Known Activations