INDEX

Explanations

words related to physical harm or injury

oai_token-act-pair · gpt-3.5-turbo

terms related to physical trauma and injury

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

assadors

-0.81

soDeliveryDate

-0.71

nostic

-0.71

Reviewer

-0.62

Alias

-0.61

issance

-0.60

Council

-0.60

ctuary

-0.59

APD

-0.58

 trustees

-0.57

POSITIVE LOGITS

 seams

0.67

 sidew

0.62

 vomit

0.61

 oxygen

0.61

 seaw

0.59

 dehyd

0.59

 drown

0.58

 undet

0.57

 centrif

0.57

 bould

0.57

Activations Density 1.855%

words related to physical harm or injury

terms related to physical trauma and injury

No Comments

No Known Activations