INDEX

Explanations

phrases referring to noticing or observing something

oai_token-act-pair · gpt-3.5-turbo

references to perceptions or observations

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

angan

-0.72

eta

-0.69

Ĥİ

-0.66

inion

-0.65

ufact

-0.63

eros

-0.63

 ensued

-0.61

aday

-0.61

 Recomm

-0.60

}{

-0.59

POSITIVE LOGITS

 VIDEOS

0.92

 therapist

0.84

 unfold

0.76

 fruition

0.70

 counselor

0.70

 firsthand

0.70

 sunrise

0.69

 devastation

0.68

 mete

0.68

 shrink

0.67

Activations Density 0.233%

phrases referring to noticing or observing something

references to perceptions or observations

No Comments

No Known Activations