INDEX

Explanations

phrases describing something as being known or recognized for a specific attribute or action

oai_token-act-pair · gpt-3.5-turbo

phrases indicating known entities or characteristics

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

essim

-0.67

uliffe

-0.63

case

-0.60

erson

-0.60

olicy

-0.59

ort

-0.58

umed

-0.58

ulo

-0.58

 deleg

-0.58

udo

-0.58

POSITIVE LOGITS

soType

0.84

è¦ļéĨĴ

0.75

ãĤ´ãĥ³

0.74

igraph

0.69

 è£ıè

0.68

\">

0.66

 initials

0.66

Strike

0.66

Ô

0.64

CHAT

0.64

Activations Density 0.117%

phrases describing something as being known or recognized for a specific attribute or action

phrases indicating known entities or characteristics

No Comments

No Known Activations