INDEX

Explanations

phrases related to interviews or statements by various individuals

oai_token-act-pair · gpt-3.5-turbo

expressions of emotional distress or concern

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

vertisement

-0.66

}.

-0.65

)--

-0.65

 pione

-0.61

').

-0.60

?),

-0.58

.--

-0.57

)?

-0.57

schild

-0.57

ankind

-0.56

POSITIVE LOGITS

 "âĢ¦

0.64

 Blumenthal

0.63

0.60

"[

0.56

0.54

 misunderstood

0.53

"#

0.53

 "...

0.53

 Ivanka

0.52

 consultations

0.52

Activations Density 1.831%

phrases related to interviews or statements by various individuals

expressions of emotional distress or concern

No Comments

No Known Activations