INDEX

Explanations

terms related to locations like "post office" or "Michigan."

oai_token-act-pair · gpt-3.5-turbo

references to time and continuity

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

hops

-0.56

 Despair

-0.55

Operation

-0.53

abwe

-0.52

iatus

-0.51

 Pastebin

-0.51

 Including

-0.51

 Annotations

-0.49

 Develop

-0.49

 defic

-0.48

POSITIVE LOGITS

iest

0.87

liest

0.82

ultimate

0.74

osphere

0.67

hest

0.57

 opposite

0.56

 equivalent

0.56

 closest

0.56

icter

0.55

most

0.54

Activations Density 0.749%

terms related to locations like "post office" or "Michigan."

references to time and continuity

No Comments

No Known Activations