INDEX

Explanations

terms related to investigations or inquiries

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "investigate" and its variations

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

the act of investigating

oai_token-act-pair · gpt-4-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

knife

-0.74

birth

-0.72

nown

-0.71

perm

-0.70

absolute

-0.70

vic

-0.70

opp

-0.68

yip

-0.68

VP

-0.68

aa

-0.67

POSITIVE LOGITS

 investigating

0.97

 investigation

0.96

 investigates

0.95

 investigations

0.86

 Investigation

0.84

 Investigative

0.84

 investigate

0.81

ĸļ

0.78

 probing

0.77

 discl

0.76

Activations Density 0.019%

terms related to investigations or inquiries

occurrences of the word "investigate" and its variations

the act of investigating

No Comments

No Known Activations