INDEX

Explanations

words related to investigation, examination, and scrutiny

oai_token-act-pair · gpt-3.5-turbo

words related to processes of examination, assessment, or analysis

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

 forgetting

-0.66

 sustaining

-0.65

 borrowing

-0.65

cakes

-0.62

uploads

-0.62

pite

-0.61

 maximal

-0.60

 unchecked

-0.60

 guarant

-0.59

 unification

-0.59

POSITIVE LOGITS

andum

0.79

 Probe

0.76

 Eval

0.75

VIEW

0.72

 Pens

0.70

 Aires

0.70

Device

0.70

ĳå£«

0.68

 Procedures

0.67

ostics

0.67

Activations Density 0.106%

words related to investigation, examination, and scrutiny

words related to processes of examination, assessment, or analysis

No Comments

No Known Activations