INDEX

Explanations

text related to replies, messages, or responses

oai_token-act-pair · gpt-3.5-turbo

mentions of the term "RE" in various contexts

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

glers

-0.84

olean

-0.80

atan

-0.79

otom

-0.73

ãĤ¨ãĥ«

-0.71

Ic

-0.70

imoto

-0.70

ocolate

-0.67

opolis

-0.66

 âĵĺ

-0.66

POSITIVE LOGITS

LEASE

1.31

VIEW

1.16

ACH

1.12

PRES

1.12

VOL

1.11

VEN

1.10

WARD

1.10

VE

1.07

CLAIM

1.06

FER

1.05

Activations Density 0.011%

text related to replies, messages, or responses

mentions of the term "RE" in various contexts

No Comments

No Known Activations