INDEX

Explanations

terms related to financial earnings or findings

oai_token-act-pair · gpt-3.5-turbo

financial terminology and references to earnings or financial results

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

INTER

-0.87

uberty

-0.77

PDATE

-0.76

 crawl

-0.75

OAD

-0.73

 exha

-0.70

REDACTED

-0.68

IRO

-0.67

 circulation

-0.67

ULTS

-0.66

POSITIVE LOGITS

ings

1.69

ers

1.50

ership

1.44

1.21

ingly

1.09

ors

1.06

scl

1.05

ances

1.02

ies

1.02

ables

0.99

Activations Density 0.069%

terms related to financial earnings or findings

financial terminology and references to earnings or financial results

No Comments

No Known Activations