INDEX

Explanations

words related to verification, such as 'verifies' and 'verifiable'

oai_token-act-pair · gpt-3.5-turbo

instances of the word "verifiable" and its variations in context

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

ACP

-0.76

hower

-0.65

iances

-0.64

 Sick

-0.63

 Parsons

-0.63

bolt

-0.62

ymm

-0.62

istics

-0.61

ivities

-0.61

 Penny

-0.60

POSITIVE LOGITS

ver

3.84

Ver

1.82

Ver

1.69

VER

1.61

ver

1.34

 verification

1.14

 verify

1.08

 Vers

1.08

VER

1.08

 authenticity

1.08

Activations Density 0.015%

words related to verification, such as 'verifies' and 'verifiable'

instances of the word "verifiable" and its variations in context

No Comments

No Known Activations