INDEX

Explanations

sentences containing comparisons or evaluations of likelihood

oai_token-act-pair · gpt-3.5-turbo

phrases expressing perceptions or judgments about situations

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

isner

-0.73

 srfAttach

-0.67

ogh

-0.65

pour

-0.64

ioch

-0.64

redo

-0.64

chance

-0.63

hurst

-0.62

addin

-0.62

orously

-0.60

POSITIVE LOGITS

 blush

0.78

 innocuous

0.76

 superf

0.74

 daunting

0.69

 confusing

0.69

 confused

0.68

 differently

0.68

 acron

0.67

 slightly

0.66

INGS

0.66

Activations Density 0.084%

sentences containing comparisons or evaluations of likelihood

phrases expressing perceptions or judgments about situations

No Comments

No Known Activations